1-Oct-87 23:48:29-PDT,15627;000000000000 Mail-From: LAWS created at 1-Oct-87 23:37:30 Date: Thu 1 Oct 1987 23:32-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #225 - CPSR, Time, Boltzmann Machines, Slava Prazdny To: AIList@SRI.COM AIList Digest Friday, 2 Oct 1987 Volume 5 : Issue 225 Today's Topics: Queries - Expert/AI Work in Communication Networks & Annual Review of Computer Science & Neural Hardware & Using the ATMS to Scan Homeric Verse, Bindings - CPSR, Representation - Time, Neural Networks - Boltzmann Machines, Obituary - Slava Prazdny ---------------------------------------------------------------------- Date: 29 Sep 87 14:33:39 GMT From: cbosgd!cblpf!dtm@ucbvax.Berkeley.EDU (Dattaram Mirvke) Subject: Need references on Expert/AI work in Communication Networks. Need references/pointers to work done in the Expert systems/AI in the Network domain. I am specially interested in simulations/modeling/ behavior analysis and fault diagnosis. Please e-mail the responses to me unless the information is of general interest. If I get sufficient responses I will summarise to the net.Thanks in advance. - Datta Miruke cbosgd!cblpf!dtm cbosgd!ncpe!drm ------------------------------ Date: 30 Sep 87 19:51:42 GMT From: wucs1!grs@uunet.UU.NET (Guillermo Ricardo Simari) Subject: Annual Review of Computer Science The book "Logical Foundations of Artificial Intelligence" by M. R. Genesereth, N. J. Nilsson contains the following reference, Levesque, H., "Knowledge Representation and Reasoning", Annual Review of Computer Science, 1986 Can anyone give me information about the above journal? I cannot find it anywhere. Guillermo Simari Washington University, Department of Computer Science St. Louis, MO, 63130, U.S.A. UUCP: grs@wucs1.UUCP or ...!{ihnp4,uunet}!wucs1!grs ------------------------------ Date: 1 Oct 87 02:02:46 GMT From: munnari!mulga.oz!jayen@uunet.UU.NET (Jayen Vaghani) Subject: Want information on Neural Hardware in use I am preparing a talk for one of my honours subjects and part of the talk centres on comparing neural hardware to other architectures. I need information on what neural hardware is in use (possibly commercially available) and what it is being used for. I would also like to know why this direction was chosen as against using a more general purpose parallel architecture and modelling the neural network on that. Perhaps some feelings about whether the approach has any future would also be nice and what problems were encountered in using the system. Responses can be to the net or mailed to me. If people are interested I will summarise any personal responses to the net. Possibly someone else has already asked this question so I would be happy to hear what responses they got. Thanks in advance, Jayen. ------- UUCP: {seismo,ukc,ubc-vision,mcvax}!mulga.oz!jayen ARPA: jayen%mulga.oz@seismo.css.gov CSNET: jayen%mulga.oz@australia ------------------------------ Date: 29 Sep 87 15:18:35 GMT From: eagle!icdoc!qmc-cs!flash@ucbvax.Berkeley.EDU (Flash Sheridan) Subject: Using the ATMS to Scan Homeric Verse As a toy demo, I'm trying to use Johan deKleer's Assumption Based Truth Maintenance System to scan Homer. I'd appreciate comments. I'd also appreciate it if somebody could email me a hundred or so lines, so I don't have to type in any more. ------------------------------ Date: 1 Oct 87 22:58:08 GMT From: acornrc!rbbb@ames.arpa (David Chase) Subject: Re: Using the ATMS to Scan Homeric Verse In article <297@sequent.cs.qmc.ac.uk>, flash@ee.qmc.ac.uk (Flash Sheridan) writes: > ... to scan Homer. > I'd also appreciate it if somebody could email me a hundred or so > lines, so I don't have to type in any more. [this doesn't really belong on this list, but this is a mighty stale pointer. I am hoping that this will jog the memory of someone else on the list with more recent information.] Sometime around about 1975 (in high school) I went to a seminar at the Nat. Junior Classical League convention where someone from Dartmouth talked about feeding the Aeneid to a computer program, doing the meter, counting "et"s, etc. I was under the impression that they had other classics on line or on tape. By the way, how should I type in ancient Greek on my U.S.A. keyboard? I can use ` and ' and ~ for accents, but what about the breath marks? David Chase, Olivetti Research Center ------------------------------ Date: 29 Sep 87 22:10:56 GMT From: sdcrdcf!ism780c!jimh@hplabs.hp.com (Jim Hori) Subject: Re: Is Computer Science Science? (Funding) In article <2868@ames.arpa> eugene@pioneer.UUCP (Eugene Miya N.) writes: >be read in the latest CPSR [Computer Professionals for Social >Responsibility] Newsletter. It appears in the halls of places like can you, or anyone, post the address of this newletter? jimh ...yeah you right ........................ ------------------------------ Date: 30 Sep 87 16:28:47 GMT From: pioneer!eugene@ames.arpa (Eugene Miya N.) Subject: Re: Is Computer Science Science? (Funding) Computer Professionals for Social Responsibility (National Office) 646 Emerson St. Palo Alto, CA 94301 ------------------------------ Date: 1 Oct 87 12:31:56 GMT From: ihnp4!homxb!homxc!del@ucbvax.Berkeley.EDU (D.LEASURE) Subject: Re: Is Computer Science Science? (Funding) In article <7397@ism780c.UUCP>, jimh@ism780c.UUCP (Jim Hori) writes: > can you, or anyone, post the address of this > newletter? [CPSR] CPSR, Inc. PO Box 717, Palo Alto CA 94301 415/322-3778 $30/yr $10/yr for student -- David E. Leasure - AT&T Bell Laboratories - (201) 615-5307 ------------------------------ Date: Tue, 29 Sep 87 09:46:05 -0400 From: koomen@cs.rochester.edu Subject: Representation of Time >From: mcvax!unido!uklirb!noekel@uunet.uu.net >Subject: J.F.Allen's work on time - (nf) Reference intervals, automatic interval hierarchy structuring, duration logic, etc, have indeed been implemented, in support of my PhD research project. For a description, watch for a UofR Tech Report and my dissertation, both expected to appear within the next year. -- Hans EMail: Koomen@CS.Rochester.Edu Paper: Johannes A. G. M. Koomen Dept. of Computer Science Phone: (716) 275-9499 [work] University of Rochester (716) 442-4836 [home] Rochester, NY 14627 ------------------------------ Date: Tue, 29 Sep 87 13:13:25 PDT From: ladkin@kestrel.ARPA (Peter Ladkin) Subject: Re: J.F. Allen's work on time There has been much subsequent work done on this topic. I'll try to summarise what I know about. James Allen's system can be formulated as a relation algebra in the sense of Tarski, and also as a complete, countably categorical first-order theory. The theory is thus decidable, and admits of quantifier elimination. This means that arbitrary first-order constraints expressed in the system are computably equivalent to a set of constraints without quantifiers. Such collections of Boolean constraints may be checked for consistency either by an extension of Allen's method, or in other ways. Marc Vilain and Henry Kautz showed that the general constraint satisfaction problem for this system is NP-complete. Allen and Pat Hayes have formulated an alternative theory of time intervals as a collection of first-order axioms. They want to allow the collection of pairs of integers as a model, and thus the axioms are weaker than the original system. These axioms have as models exactly sets of pairs from an unbounded linear order. Johan Van Benthem has investigated interval theories in his book `The Logic of Time'. All the theories are comparable, it turns out, since the collections of primitives are interdefinable. For a reading list, the AAAI-86, AAAI-87, IJCAI-85 and IJCAI-87 conference proceedings contain papers on interval systems for time representation, Allen and Hayes have a technical report (University of Rochester) due out any day, I have also technical reports not in the above sources (Kestrel Institute), and Edward Tsang (University of Essex) has some also. Tom Dean (Brown University) is using interval representations in his planner, and has investigated the most commonly occurring constraint satisfaction problems in detail. Richard Pelavin has incorporated Allen's interval system into the design for a planner, and Henry Kautz has also investigated the use of interval specifications in general planning problems. (Both are former students of James Allen). There is van Benthem's book, and a review of it by Steven Kuhn in the September 1987 Journal of Symbolic Logic (of the open problems mentioned by Kuhn, the first and last were solved already, by Roger Maddux and I, and I'm sure some others). Joe Halpern and Yoav Shoham formulated a modal logic of time with interval modalities, in the First Logic in Computer Science conference (1986, Proceedings published by IEEE). Klaus Schultz at Tubingen has a technical report comparing Allen's approach with Kamp's event theory, and Austin Tate and Colin Bell have investigated the use of interval constraints in the O-Plan planner at Edinburgh (Bell is at the University of Iowa). Other references may be found by taking the transitive closure of the `references' relation on these sources. There is very closely related work being done by Robert Kowalski's group on event structures (Imperial College). Apologies to those whose work I've missed or are unaware of (please let me know). Things are progressing fast, so we all ought to be on each other's mailing lists. peter ladkin ladkin@kestrel.arpa ------------------------------ Date: Tue, 29 Sep 1987 11:09 EDT From: "Scott E. Fahlman" Subject: Boltzmann Machines To answer your question about Boltzmann machines: In the original Boltzmann Machine formulation, a pattern (think of this as both inputs and outputs) is clamped into the visible units during the teaching phase; the network is allowed to free-run, with nothing clamped, during the normalization phase. The update of each weight is a function of the difference between co-occurrence statistics measured across that connection during the two phases. The result (if all goes well) is a trained network that has no concept of input and output: clamp a partial pattern into the visible units, and the network will try to complete it in a way that is consistent with the training examples. Clamp nothing, and the network should settle into states whose distribution approximates the distribution of examples in the training set. Later, someone (Geoff Hinton, I think), realized that if the network was really being trained to produce a certain input-to-output mapping, it was wasteful of links and training effort to train the network to reproduce the distribution of input vectors; an input will always be supplied when the network is performing. If the visible units are divided into an input set and an output set, if the teaching phase is done as before, and if the inputs (only) are clamped during the normalization phase, the network will "concentrate" on learning to produce the desired outputs, given the inputs, and will not develop the capability of reproducing the input distribution. Some papers refer to the "completion" model, others to the "Input/Ouput" model. The distinction is not always emphasized. The learning procedure is essentially the same in either case. Note that, unlike Boltzmann, the back-propagation model is inherently an I/O model, though it is possible to do completion tasks with some added work. For example, one might train a backprop network to map each of a set of patterns into itself, and then feed it partial patterns at the inputs. -- Scott Fahlman, CMU ------------------------------ Date: 30 Sep 87 21:21:50 GMT From: giraffe..arpa!krulwich@uunet.uu.net (Bruce Krulwich) Reply-to: yale.ARPA!krulwich@uunet.uu.net (Bruce Krulwich) Subject: Re: Boltzmann Machine > Since the expression for dG/dWij is the same in both cases, the > definitions of Pij- must be equivalent. The only explanation I could > think of was that "clamping" the inputs ONLY was the same thing as letting > the environment have a free run of them, so the case being described is > the free-running one. The point is that for any given inputs learning is done by comparing the desired outputs with the outputs computed by the machine. This called monitored learning, and is similar in this sense to back propogation learning. This is used for networks that perform a computation based on some input being clamped in the input units. When the output units are clamped, the P values are something like what they "should" be, so comparing these to the P values for unclamped output units lets you approximate the error between the units in qestion and learn from it. Bruce Krulwich ARPA: krulwich@yale.arpa If you're right 95% of the time, or krulwich@cs.yale.edu why worry about the other 3% ?? Bitnet: krulwich@yalecs.bitnet UUCP: {harvard, seismo, ihnp4}!yale!krulwich ------------------------------ Date: 29 Sep 87 11:04:18 PDT (Tue) From: baird@cel.fmc.com (Michael Baird) Subject: Slava Prazdny (Bindings) As many of you know by now, Slava Prazdny died Saturday, September 19th, in a hang-gliding accident, high in the California mountains. He is survived by his wife, Dagmar Dolan, their as yet unborn child, and his 15 year old daughter Bronja Prazdny. Slava was 38. He was with FMC's Santa Clara AI Center during the past two years, and before that at Schlumberger's Palo Alto Research Center / Fairchild Laboratory for Artificial Intelligence Research. Memorial Services will be held at 3:15 p.m., Wednesday October 7th, 1987, outdoors in Foothills Park, operated by the City of Palo Alto, just a few miles up Page Mill Road "west" of I-280. Flowers may be brought to the memorial. Dagmar invites members of the AI community to attend. It is suggested that you enter the park gate (it says for Palo Alto residents only -- but tell the guard that you are attending the memorial) around 3 p.m. From the parking area find the "Lee" grove (two large redwoods) beyond the picnic tables. Services will be informal, as Slava would have wanted them to be. Slava had published over 60 refereed papers, and was recognized internationally as an expert in many aspects of human and machine perception. His latest works in stereo vision and adaptive "neural" networks were deemed scientific breakthroughs. A beautiful Redwood tree in Big Basin State Park will be dedicated in Slava's name. This will be a pleasant place we can go to remember Slava. The family has asked that donations be sent in his name to The Sempirvirens Fund, 2483 Old Middlefield Way, Mountain View, CA 94043. Mike Baird baird@cel.fmc.com (408) 289-4932 ------------------------------ End of AIList Digest ******************** 1-Oct-87 23:51:34-PDT,22452;000000000000 Mail-From: LAWS created at 1-Oct-87 23:46:34 Date: Thu 1 Oct 1987 23:41-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #226 - Philosophy of AI and Computer Science To: AIList@SRI.COM AIList Digest Friday, 2 Oct 1987 Volume 5 : Issue 226 Today's Topics: Comments - Goal of AI & Nature of Computer Science ---------------------------------------------------------------------- Date: 28 Sep 87 14:36:36 GMT From: nbires!isis!csm9a!bware@ucbvax.Berkeley.EDU (Bob Ware) Subject: Re: Goal of AI: where are we going? >We all admit that the human mind is not flawless. Bias decisions >can be made due to emotional problems, for instance. ... The above has been true for all of recorded history and remains true for almost everyone today. While almost everyone's mind is flawed due to emotional problems, new data is emerging that indicates the mind can be "fixed" in that regard. To see what I am referring to, read L Ron Hubbard's book on "Dianetics". MAIL: Bob Ware, Colorado School of Mines, Golden, Co 80401, USA PHONE: (303) 273-3987 UUCP: hplabs!hao!isis!csm9a!bware or ucbvax!nbires!udenva!csm9a!bware ------------------------------ Date: 29 Sep 87 17:25:55 GMT From: eugene@pioneer.arpa (Eugene Miya N.) Reply-to: eugene@pioneer.UUCP (Eugene Miya N.) Subject: Re: Is Computer Science Science? Or is it Art? [sort of hope not] In article <8709290724.AA10633@ucbvax.Berkeley.EDU> solar!shf (Stuart Ferguson) writes: >+-- cdfk@hplb.CSNET (Caroline Knight) writes: >| ... I believe that in software there is a better analogy with art >| and illustration than engineering or science. I have noticed that this >| is not welcomed by many people in computing but this might be because >| they know so little of the thought processes and planning that go on >| behind the development of, say, a still life or an advertising poster. > >This line of thinking appeals to me alot (and I'm a "person in computing," >having 10+ years programming experience). I can apreciate this article >because my own thinking has led me to somewhat the same place regarding >"Computer Science." I'm glad I waited a bit on this. Two years ago, I met Nico Habermann of CMU. At that time I suggest CS could learn more from cognitive sciences (psychology). Habermann has an EE PhD. He didn't like this idea due to the softness. I suggest others try this question on other hard CS-types. I only ask that you avoid analogies to introspection. While the art analogy to computing has a certain appeal, especially the iterative and prototypical aspects, and it also has Knuth behind it, it also has some problems. Rather than mentioned them, I suggest you send mail to DEK and report back. From the Rock of Ages Home for Retired Hackers: --eugene miya NASA Ames Research Center eugene@ames-aurora.ARPA "You trust the `reply' command with all those different mailers out there?" "Send mail, avoid follow-ups. If enough, I'll summarize." {hplabs,hao,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene ------------------------------ Date: 29 Sep 87 17:59:04 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: Goal of AI: where are we going? In article <178@usl>, khl@usl (Calvin K. H. Leung) writes: > Should the ultimate goal of AI be the perfecting of human intel- > ligence, or the imitating of intelligence in human behavior? > > We all admit that the human mind is not flawless... So there is > no point trying to imitate the human thinking process. Some > current research areas (neural networks, for example) use the > brain as the basic model. Should we also spend some time on the > investigation of some other models which could be more efficient > and reliable? I always thought there were several different currents going in AI. One stream is trying to learn how the human mind works and imitate it. Another stream is trying to fill in the gaps in the capabilities of the human mind by using unique machine capabilities in combination with imitations of the mind. Some people are working with research objectives, some have application objectives. We don't need a unique goal for AI. We contain multitudes. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 29 Sep 87 18:25:36 GMT From: nysernic!rpicsb8!csv.rpi.edu!franklin@rutgers.edu (W. Randolph Franklin ( WRF )) Subject: Re: Is Computer Science Science? (Funding) In article <2868@ames.arpa> eugene@pioneer.UUCP (Eugene Miya N.) writes: >Status Quo? Hopefully a short note: >The reason why you have to make some clear distinctions care partially >be read in the latest CPSR [Computer Professionals for Social >Responsibility] Newsletter. It appears in the halls of places like >Ames, JPL, DOE Labs, the NAS (Natl. Acad. Sci), NSF, etc. Basically if >you are not a science, you don't get funding from those Science >Agencies. > >This is a difference in Geography (seen as an art) and Geology. >I studied remote sensing for several years. The fact that it was in a >geography --->cartography -->graph --> "art" department was a big >minus. RS is pretty respectable in some circles, and like AI, disreputable This may be improving. NSF is soliciting proposals to set up a center for excellence in Geographic Information Systems. Wm. Randolph Franklin Preferred net address: Franklin@csv.rpi.edu Alternate net: wrf@RPITSMTS.BITNET Papermail: ECSE Dept, Rensselaer Polytechnic Institute, Troy NY, 12180 Telephone: (518) 276-6077 Telex: 6716050 RPI TROU -- general RPI telex number. Wm. Randolph Franklin, RPI, 6026 JEC, (518) 276-6077, Franklin@csv.rpi.edu ------------------------------ Date: 30 Sep 87 02:08:00 GMT From: munnari!comp.vuw.ac.nz!lindsay@uunet.uu.net (Lindsay Groves) Subject: Re: Is Computer Science Science? In article <5068@jade.BERKELEY.EDU> ed298-ak@violet.berkeley.edu (Edouard Lagache) writes: >>> >.... Does Computer Science have any laws? >>> >>"Anything that can go wrong will go wrong." >> ... > > Hey those aren't laws from Computer Science, they are from the > Science (Religion?) of Murphyology.! > > E.L. The August issue of the Communications of the ACM contains an article by C.A.R.Hoare and eight others, entitled "Laws of Programming". One of their laws (4) is: ABORT U P = ABORT where ABORT (which they denote by an upside down T) is a statement that can do anything ("It places no constraint on the executing machine, which may do anything, or fail to do anything; in particular, it may fail to terminate"), and U is nondeterministic choice. The text explaining this law says: "This law is sometimes known as Murphy's Law, which state, "If it can go wrong it will"; the left-hand side describes a machine that CAN go wrong (or can behave like P), whereas the right-hand side might be taken to describe a machine that WILL go wrong. But the true meaning of the law is actually worse than this: The program ABORT will not always go wrong -- only when it ismost disastrous for it to do so! THe abundance of empirical evidence for law (4) suggests that it should be taken as the first law of computer programming." It seems that being part of "Murphyology" doesn't preclude something from being a law of Computer Science -- this one is given a very precise statement and interpretation as a law of programming, which must also count as a law of Computer Science. Given that Computer Science draws heavily on such fields as mathematics, logic, linguistics (Chomsky's hierarchy has far more relevance to Computer Science than it does to lingusitics!), electrical engineering etc., it is not surprising that laws in Computer Science should bear similarity to laws in other areas. Lindsay Groves Logic programmers' theme song: "The first cut is the deepest" ------------------------------ Date: 30 Sep 87 17:42:21 GMT From: uwslh!lishka@speedy.wisc.edu (Christopher Lishka) Subject: Re: Goal of AI: where are we going? ***Warning: FLAME ON*** In article <549@csm9a.UUCP> bware@csm9a.UUCP (Bob Ware) writes: >>We all admit that the human mind is not flawless. Bias decisions... ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ The expression "we all" does not apply to me, at very least. Some of us (at least myself)like to believe that the human mind should not be considered to be either flawed or flawless...it only "is." I feel that making a judgement on whether or not everyone admits that the human mind is flawed happens to be a biased decision on the above net-reader's part. Realize that not everyone has the same views as the above... >>...can be made due to emotional problems, for instance. ... ^^^^^^^^^^^^^^^^^^ Is this statement to be read as "emotional problems can cause bias decisions, which are flaws in the human mind?" If it does, then I heartily disagree, because I once again feel that emotional problems and/or bias decisions are not indicative of flaws in the human mind...see above for my reasons. > >The above has been true for all of recorded history and remains true >for almost everyone today. While almost everyone's mind is flawed due ^^^^^^^^^^ >to emotional problems, new data is emerging that indicates the mind can... ^^^^^^^^^^^^^^^^^^^^^ Again, I don't feel that my mind is "flawed" by emotional problems. To me that seems to be a very "Western" (and I am making a rather stereotyped remark here) method of thinking. As I have grown up with parents who have Buddhist values and beliefs, I think that making a value judgement such as "human minds are flawed because of..." should be indicated as such...there is no way to prove that sort of "fact." For all I know or care, the human mind is neither perfect nor flawed; it just "is," and I don't wish to make sweeping generalities such as the above. There are many other views of the mind out there, and I recommend looking into *all* Religious views as well as *all* Scientific views before even attempting a statement like the above (which would easily take more than a lifetime). >...be "fixed" in that regard. To see what I am referring to, read L Ron ^^^^^ >Hubbard's book on "Dianetics". To me this seems to be one of many problems in A.I.: the assumption that the human mind can be looked at as a machine, and can be analyzed as having flaws or not, and subsequently be fixed or not. That sort of thinking in my opinion belongs more in ones Personal Philosophy and probably should not be used in a "Scientific" (ugghh, another hard-to-pin-down word) argument, because it is damned hard to prove, if it is able to be proven at all. I feel that the mind just "is," and one cannot go around making value judgements on another's thoughts. Who gives anyone else the right to say a person's mind is "flawed?" To me that kind of judgement can only be made by the person "owning" the mind (i.e. who is thinking and communicating with it!), and others should leave well enough alone. Now I realize that this brings up arguments in other fields (such as Psychology), but I feel A.I. should try and move away from these sort of value judgements. A comment: why don't A.I. "people" use the human mind as a model, for better or for worse, and not try to label it as "flawed" or "perfect?" In the first place, it is like saying that something big (like the U.S. Government) is "flawed;" this kind of thing can only be proven under *certain*conditions*, and is unlikely to hold for all possible "states" that the world can be in. In the second place, making that kind of judgement would seem to be fruitless given all that we *do*not* know about the human brain/mind/soul. It seems to me to be like saying "hmmmm, those damned quarks are fundamentally flawed", or "neuronal activity is primarily flawed in the lipid bilayer membrane." I feel that we as humans just do not know diddley about the world around us, and to say it is flawed is a naive statement. Why not just look at the human mind/brain as something that has evolved and existed over time, and therefore may be a good model for A.I. techniques UNDER CERTAIN CIRCUMSTANCES? A lot less people would be offended... ***FLAME*OFF*** Sorry if the above offends anyone...but the previous remarks offended me enough to send a followup message around the world. If one is going to make remarks based on very personal opinions, try to indicate that they are such, and please remember that not everyone thinks the way you do. Of course, pretty much everything I said above is a personal opinion, and I don't presume that even one other person thinks the same way as I do (but it would be nice to know that others think similarily ;-). Disclaimer: the above views are my thoughts only, and do not reflect the views of my employer, although there is eveidence that my cockatiels are controlling my thoughts !!! ;-) -Chris -- Chris Lishka /lishka@uwslh.uucp Wisconsin State Lab of Hygiene <-lishka%uwslh.uucp@rsch.wisc.edu \{seismo, harvard,topaz,...}!uwvax!uwslh!lishka ------------------------------ Date: 30 Sep 87 22:09:09 GMT From: topaz.rutgers.edu!josh@rutgers.edu (J Storrs Hall) Subject: Re: Goal of AI: where are we going? lishka@uwslh.UUCP (Christopher Lishka) writes: In article <549@csm9a.UUCP> bware@csm9a.UUCP (Bob Ware) writes: >>We all admit that the human mind is not flawless. Bias decisions... ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ The expression "we all" does not apply to me, at very least. Some of us (at least myself)like to believe that the human mind should not be considered to be either flawed or flawless...it only "is." It seems to me that this simply means that you hold the words "flawed" and "flawless" to be meaningless. It is as if Bob Ware were saying that the human mind were not plegrontless. Only I don't see why I would get so upset if I saw people saying that minds are plegronted at best, even if I didn't understand what they meant by the term. I would instead make an effort to comprehend the concepts being used. >>...can be made due to emotional problems, for instance. ... Is this statement to be read as "emotional problems can cause bias decisions, which are flaws in the human mind?" If it does, then I heartily disagree, because I once again feel that emotional problems and/or bias decisions are not indicative of flaws in the human mind...see above for my reasons. I would say that an emotional *problem* is by definition a flaw. If you believe that Manson and Hitler and Caligula were not flawed, but that is just the "way they were", and there is no reason to prefer Thomas Aquinas over Lyndon LaRouche, then your own reasoning is distinctly flawed. To me that seems to be a very "Western" (and I am making a rather stereotyped remark here) method of thinking. As I have grown up with parents who have Buddhist values and beliefs, I think that making a value judgement such as "human minds are flawed because of..." should be indicated as such...there is no way to prove that sort of "fact." Can you say "evangelical fundamentalist mysticism"? Your Eastern values seem to be flavored by a strong Western intellectual aggressiveness, which seems contradictory. Twice the irony in a pound of holy calves liver. There are many other views of the mind out there, and I recommend looking into *all* Religious views as well as *all* Scientific views before even attempting a statement like the above (which would easily take more than a lifetime). What an easy way to sidestep doing any real thinking. Do you suggest that we should read all the religious writings having to do with angels before we attempt to build an airplane? Do you think that one must be an expert on faith healing and the casting out of demons before he is allowed to make a statement about this interesting mold that seems to kill bacteria? In Western thought it has been realized at long and arduous last that the appeal to authority is fallacious. Experiment works; the real world exists; objective standards can be applied. Even to people. >...be "fixed" in that regard. To see what I am referring to, read L Ron >Hubbard's book on "Dianetics". Experiment (the church of scientology) shows that Hubbards ideas in this regard are hogwash. Hubbard's phenomenon had much more to do with the charismatic religious leaders of the past, than the rational enlightenment of the future. To me this seems to be one of many problems in A.I.: the assumption that the human mind can be looked at as a machine, and can be analyzed as having flaws or not, and subsequently be fixed or not. Surely this is independent of the major thrust of AI, which is to build a machine that exhibits behaviors which, in a human, would be called intelligent. It is true that most AI researchers "believe that the mind is a machine", but it seems that the alternative is to suggest that human intelligence has a supernatural mechanism. That sort of thinking in my opinion belongs more in ones Personal Philosophy and probably should not be used in a "Scientific" (ugghh, another hard-to-pin-down word) argument, because it is damned hard to prove, if it is able to be proven at all. My personal philosophy *is* scientific, thank you, and it is an objectively better one than yours is. I feel that the mind just "is," and one cannot go around making value judgements on another's thoughts. Who gives anyone else the right to say a person's mind is "flawed?" Who gives me the right to say that 2+2=4 when you feel that it should be 5? If the Wisconsin State Legislature passed a law saying that it was 5, they would be wrong; if everybody in the world believed it was 5, they would be wrong; if God Himself claimed it was 5, He would be wrong. A comment: why don't A.I. "people" use the human mind as a model, for better or for worse, and not try to label it as "flawed" or "perfect?" In the first place, it is like saying that something big (like the U.S. Government) is "flawed;" this kind of thing can only be proven under *certain*conditions*, and is unlikely to hold for all possible "states" that the world can be in. But the U.S. Government IS flawed... In the second place, making that kind of judgement would seem to be fruitless given all that we *do*not* know about the human brain/mind/soul. Back in the middle ages, we didn't know much about the Black Plague, but it was obvious that someone who caught it became pretty flawed pretty fast. Furthermore, this small understanding was considered sufficient grounds to inflict the social snubs of not associating with such a person. It is incredibly arrogant to declare that we must not make any judgements until we know everything. The whole point of having a human mind rather than a rutabaga is that you *are* able to make judgements in the absence of complete information. Brains evolving in a natural setting have always had to make *life-and-death* decisions on the spur of the moment with whatever information was available. Is that large furry creature dangerous? You've never seen a grizzly bear before. No time to consult the views of all the world's ancient religions on the subject... I feel that we as humans just do not know diddley about the world around us, and to say it is flawed is a naive statement. To say that it is not flawed is just simply idiotic. If you apply enough sophistry you may manage to get the conversation to a level where the original statement is meaningless. For example, there are (or may be) no "flawed" atoms in a broken radio. But to change the level of discussion as a rhetorical device is tantamount to lying. To do it without realizing you are doing it is tantamount to gibberish. Sorry if the above offends anyone... It offends me greatly. The anti-scientific mentality is an emotional excuse used to avoid thinking clearly. It would be much more honest to say "I don't want to think, it's too hard work." Can't you see the contradiction involved in criticizing someone for exercising his judgement? The champions of irrationality, mysticism, and superstition have emotional problems which bias their cognitive processes. Their minds are flawed. --JoSH ------------------------------ Date: 30 Sep 87 16:31:00 GMT From: uxc.cso.uiuc.edu!osiris.cso.uiuc.edu!goldfain@a.cs.uiuc.edu Subject: Re: Goal of AI: where are we going? Bob Ware, of Colorado School of Mines, writes : > ... While almost everyone's mind is flawed due to emotional problems, new > data is emerging that indicates the mind can be "fixed" in that regard. To > see what I am referring to, read L Ron Hubbard's book on "Dianetics". I suppose that if someone feels they have emotional problems and turned to Mr. Hubbard for help, there is some sense to that. He ought to know about them, since reports have indicated over the years that he has more than his fair share of them ... :-) Alternatively, one could consult someone who actually has credentials in psychology. "You pays your money and you takes yer choice." - Mark Goldfain (ARPA: goldfain@osiris.cso.uiuc.edu) ------------------------------ End of AIList Digest ******************** 4-Oct-87 23:53:02-PDT,19233;000000000000 Mail-From: LAWS created at 4-Oct-87 23:43:22 Date: Sun 4 Oct 1987 23:41-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #227 - Goal of AI To: AIList@SRI.COM AIList Digest Monday, 5 Oct 1987 Volume 5 : Issue 227 Today's Topics: Philosophy - Goal of AI ---------------------------------------------------------------------- Date: Sun, 4 Oct 87 14:56:42 -0200 From: Jacob Levy Subject: Another Blooming Endless Argument Please! It seems we are getting ready for another deluge, this time under the title of "Re: Goal of AI: where are we going?". While the subject line certainly does justify its inclusion in AI-digest, it may easily get out of hand and deteriorate into personal arguments. There are already first signs of super heated discussions and personal attacks in V5 #226. The question I am asking myself when reading these postings is "how much of this material is AI-related, and how much is purely philosophical/psycholo- gical or whatever?" I suggest that authors participating in this discussion would benefit from application of the same criterion - how much is this an ARTIFICIAL intelligence-related posting? The word ARTIFICIAL is crucial and determines, for me at least, whether I want to read on. Another Please! Remember that this is my personal opinion, I am "a small egg" only, so no flames or personal attacks. Educate, not eradicate, OK? P.S. Is the discussion entitled "Nature of Computer Science" really appro- priate for AIlist? Rusty Red (AKA Jacob Levy) BITNET: jaakov@wisdom ARPA: jaakov%wisdom.bitnet@wiscvm.wisc.edu CSNET: jaakov%wisdom.bitnet@relay.cs.net ------------------------------ Date: 1 Oct 87 18:36:20 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: Goal of AI: where are we going? In article <270@uwslh.UUCP>, lishka@uwslh.UUCP (Christopher Lishka) writes: > ***Warning: FLAME ON*** > > In article <549@csm9a.UUCP> bware@csm9a.UUCP (Bob Ware) writes: > >>We all admit that the human mind is not flawless. Bias decisions... > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > The expression "we all" does not apply to me, at very least. Some of > us (at least myself)like to believe that the human mind should not be > considered to be either flawed or flawless...it only "is." .... Some interesting points here. Point one, the human mind is in fact a phenomenon, and phenomena are neither flawed nor perfect, they are the stuff that observation is made of. Score one for Lishka. Point two, we keep using the human mind as a tool, to solve problems. As such, it is not merely a phenomenon, but a means to an end, and is subject to judgments of its utility for that purpose. Now we can say whether it is perfect or flawed. Obviously, it is not perfect, since we often make mistakes when we use it. Score one for Ware. Point three, when we try to make better tools, or tools to supplement the human mind, all these improvements are created by the human mind. In fact, the purposes of these tools are created by the human mind. The human mind is thus the ultimate reasoning tool. Score one for the human mind. You might say the same of the human hand. As a phenomenon, it exists. As a tool, it is imperfect. And it is the ultimate mechanical tool, since all mechanical tools are directly or indirectly made by it. It is from these multiple standpoints that we derive the multiple goals of AI: to study the mind, to supplement the mind, and to serve the mind. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 1 Oct 87 13:22:00 GMT From: uxc.cso.uiuc.edu!uxe.cso.uiuc.edu!morgan@a.cs.uiuc.edu Subject: Re: Goal of AI: where are we going? Maybe you should approach it as a scientist, rather than an engineer. Think of the physicists: they aren't out to fix the universe, or construct an imitation; they want to understand it. What AI really ought to be is a science that studies intelligence, with the goal of understanding it by rigorous theoretical work, and by empirical study of systems that appear to have intelligence, whatever that is. The best work in AI, in my opinion, has this scientific flavor. Then it's up to the engineers (or society at large) to decide what to do with the knowledge gained, in terms of constructing practical systems. ------------------------------ Date: 2 Oct 87 15:31:04 GMT From: bloom-beacon!gatech!udel!montgome@husc6.harvard.edu (Kevin Montgomery) Subject: Re: Goal of AI: where are we going? In article <259@tut.cis.ohio-state.edu> tanner%tut.cis.ohio-state.edu@osu-eddie.UUCP (Mike Tanner) writes: >If you want to say that what I'm doing is not AI, fine. I think it is, but if >you'll give me a better name I'll take it and leave AI to the logicians. It >is not psychology (my experiments involve building programs and generally >thinking about computational issues, not torturing college freshmen). And I'm >not really interested in duplicating the human mind, it's just that the human >mind is the only intelligence I know. Welcome to the fascinating world of Cognitive Modelling! If AI is to be pure logic, more power to it. But the "real world" usually doesn't let one work only with pure logic- the case of incomplete information is an example. If you saw me driving towards work, and it is around 9am, you may conclude that I'm going to work. However, from a purely logical point of view, my direction of travel has little to do with my end destination. Whatever. I think modelling's more fun anyway! (no flames about AI handling the above situation and 'most-probable scenario' stuff, please) -- Kevin Montgomery ------------------------------ Date: 1 Oct 87 12:49:39 GMT From: tanner@tut.cis.ohio-state.edu (Mike Tanner) Reply-to: tanner%tut.cis.ohio-state.edu@osu-eddie.UUCP (Mike Tanner) Subject: Re: Goal of AI: where are we going? In article <270@uwslh.UUCP> lishka@uwslh.UUCP (Christopher Lishka) writes: >In article <549@csm9a.UUCP> bware@csm9a.UUCP (Bob Ware) writes: >>We all admit that the human mind is not flawless. Bias decisions... > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > [the underscored bit above indicates a number of faulty assumptions, > e.g., that it makes sense to talk about "flaws" in the mind.] > I liked this reply. Whether the problem is "western" philosophy or not, I'm not sure. It may be true for the casual AI dabbler. I.e., the average intelligent person on first thinking or hearing of the topic of AI will often say things like, "But people make mistakes, do you really want to build human-like machines?" Within AI itself this attitude manifests itself as rampant normativism. Somebody adopts a model of so-called correct reasoning, e.g., Bayesian decision theory, logic, etc., and then assumes that the abundant empirical evidence that people are unable to reason this way shows human reasoning to be flawed. These people want to build "correct" reasoning machines. I say, OK, go ahead. But that's not what I want to do. I want to understand thinking, intelligent information processing, problem-solving, etc. And I think the empirical evidence is trying to tell us something important. I am not sure just what. It seems clear that thinking is not logical (which is not to say "flawed" or "incorrect", merely "not logical"). An interesting question is, "why not?" People are able to use language, solve problems -- to think -- but is that in spite of illogic or because of it or neither? I don't think we're going to understand intelligence by adopting an a priori correct model and trying to build machines that work that way (except by negative results). If you want to say that what I'm doing is not AI, fine. I think it is, but if you'll give me a better name I'll take it and leave AI to the logicians. It is not psychology (my experiments involve building programs and generally thinking about computational issues, not torturing college freshmen). And I'm not really interested in duplicating the human mind, it's just that the human mind is the only intelligence I know. -- mike tanner Dept. of Computer and Info. Science tanner@ohio-state.arpa Ohio State University ...cbosgd!osu-eddie!tanner 2036 Neil Ave Mall Columbus, OH 43210 ------------------------------ Date: 3 Oct 87 06:14:38 GMT From: vax1!czhj@cu-arpa.cs.cornell.edu (Ted Inoue) Subject: Re: Goal of AI: where are we going? (Where should we go...) In article <46400008@uxe.cso.uiuc.edu> morgan@uxe.cso.uiuc.edu writes: > >Maybe you should approach it as a scientist, rather than an engineer. Think >... >What AI really ought to be is a >science that studies intelligence, with the goal of understanding it by >rigorous theoretical work, and by empirical study of >systems that appear to have intelligence, whatever that is. The best work >in AI, in my opinion, has this scientific flavor. Then it's up to the >engineers (or society at large) to decide what to do with the knowledge >gained, in terms of constructing practical systems. I wholeheartedly support this idea. I'd go even further however, and say that most "AI" research is a huge waste of time. I liken it to using trial and error methods like those used by Edison which led him to try thousands of possibilities before hitting one that made a good lightbulb. With AI, the problem is infinitely more complicated, and the chance of finding a solution by blind experimentation is nil. On the other hand, if we take an educated approach to the problem, and study 'intelligent' systems, we have a much greater chance of solving the mysteries of the mind. Some of you may remember my postings from last year where I expounded on the virtues of cognitive psychology. After investigating research in this field in more detail, I came up very disillusioned. Here is a field of study in which the soul purpose is to scientifically discover the nature of thought. Even with some very bright people working on these problems, I found that the research left me cold. Paper after paper describe isolated phenomena, then go on to present some absurdly narrow minded theory of how such phenomena could occur. I've reached the conclusion that we cannot study the mind in isolated pieces which we try to put together to form a whole. But rather we have to study the interactions between the pieces in order to learn about the pieces themselves. For example, take vision research. Many papers have been written about edge detection algorithms, possible geometries, and similarly reductionist algorithms for making sense of scenes. I assert that the interplay between the senses and the experiential memory is huge. Further, because of these interactions, no simple approach will ever work well. In fact, what we need is to study the entire set of processes involved in seeing before we can determine how we perceive objects in space. This is but a single example of the complexity of studying such aspects of the mind. I found that virtually every aspect of cognition has such problems. That is, no aspect is isolated! Because of this immensely complex set of interactions, I believe that the connectionist theories are heading in the right direction. However, these theories are somewhat too reductionistic for my tastes as well. I want to understand how the mind works at a high level (if possible). The actual implementation is the easy part. The understanding is the hard part. ---Ted Inoue ------------------------------ Date: 3 Oct 87 23:47:07 GMT From: lishka@uwslh.UUCP (Christopher Lishka) Reply-to: lishka@uwslh.UUCP (Christopher Lishka) Subject: Re: Goal of AI: where are we going? In article <46400008@uxe.cso.uiuc.edu> morgan@uxe.cso.uiuc.edu writes: > >Maybe you should approach it as a scientist, rather than an engineer. Think >of the physicists: they aren't out to fix the universe, or construct an >imitation; they want to understand it. I think this is a good point. I have always thought that Science was a method used to predict natural events with some accuracy (as opposed to guessing). Whether this is understanding, well I guess that depends on one's definition. I like this view because it (to me at least) parallels the attempts by nearly all (if not all) religions to do the same thing, and possibly provide some form of meaning to this strange world we live in. It also opens the possibility of sharing views between scientists and other people explaining the world they see with their own methods. >What AI really ought to be is a >science that studies intelligence, with the goal of understanding it by >rigorous theoretical work, and by empirical study of >systems that appear to have intelligence, whatever that is. The best work >in AI, in my opinion, has this scientific flavor. Then it's up to the >engineers (or society at large) to decide what to do with the knowledge >gained, in terms of constructing practical systems. I like this view also, and feel that A.I. might go a little further in studying other areas in conjunction with the human mind. Maybe this isn't pure A.I., but I'm not sure what pure A.I. is. One interesting note is that maybe the people who are implementing various Expert Systems (which grew out of A.I. research) for real-world applications are the "engineers" of which morgan@uxe speaks of. And more power to both the "scientists" and "engineers" then, and those in the gray area in between. It's good to be able to work together like this, and not have the "scientists" only come up with research that cannot be applied. Disclaimer: I am sitting here typing this because my girfriends cat is holding a gun at my head, and am in no way responsible for the content ;-) [If anyone really wants to flame me, please mail me; if you really think there is some benefit in posting the flame, go ahead. I reply to all flames, but if my reply doesn't get to you, it is because I am not able to find a reliable mail path (which is too damned often!)] -Chris -- Chris Lishka /lishka@uwslh.uucp Wisconsin State Lab of Hygiene <-lishka%uwslh.uucp@rsch.wisc.edu \{seismo, harvard,topaz,...}!uwvax!uwslh!lishka ------------------------------ Date: 1 Oct 87 09:11:37 GMT From: mcvax!enea!kuling!waldau@uunet.uu.net (Mattias Waldau) Subject: Re: Goal of AI: where are we going? In article <178@usl> khl@usl.usl.edu.UUCP (Calvin Kee-Hong Leung) writes: >Provided that we have the necessary technology to build robots >that are highly intelligent; they are efficient and reliable and >they do not possess any "bad" characteristic that man has. Then >what will be the roles man plays in the society where his intel- >ligence can be viewed as comparatively "lower form"? > One of the short stories in Asimov's "I, robot" is about the problem mentioned in the previous paragraph. It is about a robot and two humans on a space station near our own sun. I can not tell more, otherwise I spoil your fun. It is very good! ------------------------------ Date: Sun, 4 Oct 87 20:54:01 -0200 From: Eyal mozes Subject: Re: Goal of AI: Where are we Going? > I believe that those "bad" characteristics of human are necessary > evils to intelligence. For example, although we still don't understand > the function of emotion in human mind, a psychologist Toda saids that > it is a device for servival. When an urgent danger is approaching, you > don't have much time to think. You must PANIC! Emotion is a meta- > inference device to control your inference mode (mainly of recources). > > If we ever make a really intelligent machine, I bet the machine > also has the "bad" characteristics. In summary, we have to study > why human has those characteristics to understand the mechanism of > intelligence. I think what you mean by "the bad characteristics" is, simply, free will. Free will includes the ability to fail to think about some things, and even to actively evade thinking about them; this is the source of biased decisions and of all other "flaws" of human thought. Emotions, by themselves, are certainly not a problem; on the contrary, they're a crucial function of the human mind, and their role is not limited to emergencies. Emotions are the result of subconscious evaluations, caused by identifications and value-judgments made consciously in the past and then automatized; their role is not "to control your inference mode", but to inform you of your subconscious conclusions. Emotional problems are the result of the automatization of wrong identifications and evaluations, which may have been reached either because of insufficient information or because of volitional failure to think. A theory of emotions and of free will, explaining their role in the human mind, was developed by Ayn Rand, and the theory of free will was more recently expanded by David Kelley. Basically, the survival value of free will, and the reason why the process of evolution had to create it, is man's ability to deal with a wide range of abstractions. A man can form concepts, gain abstract knowledge, and plan actions on a scale that is in principle unlimited. He needs some control on the amount of time and effort he will spend on each area, concept or action. But because his range his unlimited, this can't be controlled by built-in rules such as "always spend 1 hour thinking about computers, 2 hours thinking about physics" etc.; man has to be free to control it in each case by his own decision. And this necessarily implies also freedom to fail to think and to evade. It seems, therefore, that free will is inherent in intelligence. If we ever manage to build an intelligent robot, we would have to either narrowly limit the range of thoughts and actions possible to it (in which case we could create built-in rules for controlling the amount of time it spends on each area), or give it free will (which will clearly require some great research breakthroughs, probably in hardware as well as software); and in the later case, it will also have "the bad characteristics" of human beings. Eyal Mozes BITNET: eyal@wisdom CSNET and ARPA: eyal%wisdom.bitnet@wiscvm.wisc.edu UUCP: ...!ihnp4!talcott!WISDOM!eyal ------------------------------ End of AIList Digest ******************** 4-Oct-87 23:56:52-PDT,16167;000000000000 Mail-From: LAWS created at 4-Oct-87 23:53:02 Date: Sun 4 Oct 1987 23:46-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #228 - Philosophy, Native American Languages To: AIList@SRI.COM AIList Digest Monday, 5 Oct 1987 Volume 5 : Issue 228 Today's Topics: Queries - Connection Graphs & IEEE Neural Net Conference & Expert Systems Company Advice & TI Explorer/Common Lisp/PCs & Protocols, Graphics - Summary Pending, Philosophy - Flaws, Linguistics - Natural Kinds and Indians ---------------------------------------------------------------------- Date: 2 Oct 87 11:53:46 GMT From: mcvax!enea!kuling!nilsh@uunet.uu.net Subject: Connection Graphs I've recently heard that there has been quite a lot of work done on Connection Graphs in West Germany the past few years. I would like to get in touch with people who has been involved in this, especially from Munchen, Kaiserslauten and Karlsruhe. My main interest for the moment is results concerning completeness for the Connec- tion Graph Proof Procedure of Kowalski. Please contant me on net-mail or "snail-mail". Net-Mail: nilsh@kuling.UUCP Snail-Mail: Nils Hagner Dept. of Computing Science Uppsala University P.O. Box 520 751 20 Uppsala SWEDEN ------------------------------ Date: 4 Oct 87 02:54:50 GMT From: munnari!latcs1.oz.au!suter@uunet.UU.NET (David Suter) Subject: IEEE conf. (Boulder) re: "Neural Info. Proc. Systems - Natural and Synthetic" conf. Boulder, Colorado - Nov. 8-12 1987. I would like to contact the registration people quickly. My snail mail to some of the organisers has either been mis-directed or lost. Thus if anyone can supply a direct conference telephone number, or an e-mail address for registration - I would be grateful. --------------------- David Suter ISD: +61 3 479-2596 Department of Computer Science, STD: (03) 479-2596 La Trobe University, ACSnet: suter@latcs1.oz Bundoora, CSnet: suter@latcs1.oz Victoria, 3083, ARPA: suter%latcs1.oz@uunet.uu.net Australia UUCP: ...!uunet!munnari!latcs1.oz!suter TELEX: AA33143 FAX: 03 4785814 ------------------------------ Date: 30 Sep 87 17:11:06 GMT From: decvax!dartvax!waltervj@ucbvax.Berkeley.EDU (walter jeffries) Subject: Expert Systems Company Financing... I am in the process of starting a company to do expert systems developement in the field of psychiatry. Principles include two top domain experts in this field, an expert in data anlysis, and an MBA canidate with training in marketing in the computer field. I am not the business end of things but would appreciate any comments/experiences that people may have with getting capital (sources, things to be careful of, etc.). [...] Many thanx, -Waltervj@dartvax ------------------------------ Date: Fri, 02 Oct 87 19:44:19 GMT From: A385%EMDUCM11.BITNET@WISCVM.WISC.EDU Subject: TI Explorer, Common Lisp & PC's Date: 2 October 1987, 19:42:38 GMT From: A385 at EMDUCM11 To: AILIST-REQUEST at SRI Hello AI community from Spain!! We are a group of absolute beginners using the TI EXPLORER machine. Our problem is that we only have two 'explorers' for a lot of people and we'd like to profite our PC's (AT's) in order to get experience using Common Lisp, but two question arise: 1) Which is the best Common Lisp implementation (with flavors, packages....) running on AT's??. Is it Golden Common Lisp? 2) Does anyone has any experience connecting PC's and TI Explorers to transfer files? Is it posible ? Thank you very much in advance for any help or suggestion. Yours Javier Lopez ------------------------------ Date: Fri, 2 Oct 87 08:05:37 edt From: steve@hubcap.clemson.edu ("Steve" Stevenson) Subject: Query on protocols. I am interested in determining what general principles apply to defining any type (human,machine, etc) of communications protocol. As an example, what would one have to do to establish that two protocols are the "same". [i.e., message-passing vs shared memory/wait/signal]. Steve (really "D. E.") Stevenson steve@hubcap.clemson.edu Department of Computer Science, (803)656-5880.mabell Clemson University, Clemson, SC 29634-1906 ------------------------------ Date: 30 Sep 87 21:59:17 GMT From: mcvax!ukc!mupsy!mucs!arnold@uunet.uu.net (Toby Howard) Subject: Thanks for AI/Graphics... Some time ago I posted a plea asking for any refs people could give me on connecting AI and Graphics, and I promised to summarise. This note is to say I'm really grateful for all the help I've received, and I *really will* summarise, and thank people individually. Just now I'm mega-busy---but I haven't forgotten! toby [This is a shared account. Please ignore the From: field, and reply to the following address. Thanks] Toby Howard Computer Graphics Unit, Manchester University, UK. janet: thoward@uk.ac.man.cs.cgu internet: thoward%cgu.cs.man.ac.uk@nss.cs.ucl.ac.uk ------------------------------ Date: 02 Oct 87 20:26:00 EDT From: Walter Roberson Subject: flaws In AILIST of October 2nd, Christopher Lishka (uwslh!lishka@speedy.wisc.edu), and J Storrs Hall (topaz.rutgers.edu!josh@rutgers.edu) discuss whether human minds are inherently flawed. Chris proposes that human minds just *are*, neither flawed nor unflawed; JoSh disagrees strongly, and claims Chris's position to not be scientifically based. Leaving aside for the moment the question of whether mathematics is a science (at last notice, AILIST list hadn't resolved that one), I believe that I can offer a mathematical basis for Chris's position. Consider a set (possibly infinite) of objects, U, and at least two single-place predicates over that set, P, and Q. Add n-ary predicates and distinguished constants, if you like. Consider the following first-order sentance over this language: "For all x in U, Px => Qx". Is this sentance true? It depends on the relations P and Q. If Qx is "false" for all x in U, and at least one Px is "true", then the sentance is false -- for that P and Q. If Px is "false" for all x in U, then the sentance is true -- again, for that P and Q. Thus, the truth of the sentance depends upon the structure (U, set(P, Q, etc), set(constants), set(n-ary functions)) in which it is evaluated. Now, as there is at least one such structure in which this sentance is false, the sentance is NOT "logically implied" by the language of its formulation. And, as there is also at least one such structure in which the sentance is true, one can only talk about the validity of the sentance in terms of its value in a particular structure. Loosely speaking, the validity of the sentance varies with the interpretation one gives to the relationships. Consider now the above sentance, ("for all x in U, Px => Qx") with the human intepretation that it denotes "all minds are flawed" -- that is, Px being interpreted as the predicate "x is a human mind", and Qx being intepreted as the predicate "x is flawed". Assigning the sentance a human interpretation makes it no more true or false than before: the difference is only in the emotional zing of the interpretation. Assigning a validity to the sentance based on a religious set of values corresponds to chosing a structure and evaluating the sentance within that structure. The sentance may be valid or invalid within that structure, but, in isolation, the sentance will still be neither true nor false. Chris's position is that "human minds are flawed" is only true within certain belief sets: that it is not a true statement because it is not a logically implied statement. JoSH's position is that the interpretations of the words "human minds", "are" and "flawed", are such that the statement is implicitly true: that semantically, the statement is automatically self-restricting to the class of structures in which it is true. Certainly the conventional wisdom is that "nobody's perfect". That has a certain intuitive "rightness" to it which is very compelling. And if nobody is perfect, then everyone is flawed, right? But what someone saying, "Nobody's perfect" really means is, "There isn't anyone that measures up to my standards of perfection". That, however, is more a reflection of the utterer's standards of perfection than upon the intrinsic qualities of any other given person. A lot of people have done things which haven't pleased me, but that's a matter of my expectations, rather than a question of whether they were "flawed" or not. --- In part of his response, JoSH disapproves of Chris's position, based upon operational grounds. Indeed, we do not -need- to study the aerodynamics of angels in order to build an airplane. I don't believe, though, that Chris implied that we needed to do so: rather, he favours a position closer to the doctrine of necessity; that if X isn't necessary in order to do Y, and Y is your goal, then don't do X. In this case, X is "assign a definite truth value to 'human minds are flawed'", and Y is "computationally model a human intelligence". Chris believes X to be unnecessary (and impossible in finite time anyways). JoSH believes it to be possible; I haven't been watching closely enough to determine whether he believes it to be necessary. --- Is 2+2=4 ? In the ring Z4, No: 2+2=0 instead. And since '=' is merely the symbol for a binary operation, traditionally a certain well-known predicate, then sometimes 2+2=5 afterall. Try, for example, reading '=' as denoting the binary predicate traditionally represented as '<'. Is the broken radio flawed? Well, if it was hit by lightning while playing "satanic rock music", and melted down into a representation of "Jesus", I rather doubt people would call it "flawed" when they couldn't get music out of it. Not much use in trying to decide whether an object is "flawed" or "bad" or "evil" or whatever -- if it doesn't do what you want it to, perhaps it'll make a dandy paperweight instead. Or bonfire fuel, if you've found it particularily frustrating. Is a dead person "flawed" because they are no longer living? I'm told that death is a very natural process -- happens to everyone, they say. But its not going to happen to me -- at least not during my lifetime! (Thanks, Raymond!) Walter Roberson --- Reference: "A Mathematical Introduction to Logic", Herbert B. Enderton, 1972, Academic Press ------------------------------ Date: Sat, 3 Oct 87 10:30:14 EDT From: Bruce Nevin Subject: natural kinds and Indians > From: cugini@icst-ecf.arpa > I believe there have been anthropological studies, for instance, > showing that Indian classifications of animals and plants line > up reasonably well with the conventional Western taxonomy. I saw this go by in AIList, and here it comes again in NL-KR, and I just can't let you get away with it, John. Glib references to `Indian classifications of animals and plants' remind one of titles in the 17th century like `The Indian Language Reduced to Grammar'. `Indian classifications', indeed! Which of the hundreds of Amerindian languages? Which of the half-dozen or so linguistic families in North America alone? Linguistic families in the Americas are as diverse from each other as the Indo-European family is from the Sino-Tibetan family, and as Finno-Ugaritic is from both: there is no demonstrated genetic relationship whatsoever. If the claim is across all Amerindian languages, it seems preposterous on the face of it. In some languages, terms for animals and plants are composite, derived from or related to predicative compounds of the type `water-strider'. In a polysynthetic language, some of the elements underlying such a compound might be classificatory morphemes that imply a rather different taxonomy. Certain of these we might gloss e.g. `long, slender object' or `spherical object' or `flexible object'. Examining our glosses for words incorporating these elements as affixes or infixes, however, we always see abundant grounds for doubting that we have captured the Indian generalization in our English net. What do `both arms', `lips', `encircle', `sew' have in common? `Soft opposed forces' is the gloss given for Pomo bi-. How about `fire, heat, cold, light, emotions, mind'? Pomo mu- is glossed `nonlong object through the air', and the above are glosses for its contribution in just some of its occurrences. In other languages, such terms are (synchronically at least) primitive, of the type `cat'. What do `horse', `dog', and `slave' have in common? All are translations of caH:o'm in Achumawi, which appears to refer to a social role rather than anything like genus or species. Indeed, all such terms in Achumawi seem to imply place in a kind of `social' structure involving all beings, a mental system orthogonal to our Realist presumptions about `objective' `external' reality. Theories of animism begin to get at it, perhaps, and here you might begin to get at some cultural/religious commonality among peoples in the Americas. In Wappo and in Yana, the word for 'dog' and `horse' is again the same, but is the Spanish loanword chucho (cu:cu' in Wappo, su:su [pronounced something like shoo-shoo] in Yana). Why not the Spanish word for horse, cavallo? I don't have any information on the Wappo and Yana words for `slave', but suspect strongly that the same `taxonomy' has a role here. Compare Wappo ka'wa:yu?+ne'w `horse-yellowjacket', perhaps on the analogy of English `horsefly', where ?ne'w is `yellowjacket'. Achumawi, Pomo, and Yana are all Northern Hokan languages, b.t.w., and are (or were) in fairly close proximity in Northern California, whereas Wappo is an unrelated Yukian language a bit further south, between the Pomo languages/dialects and San Francisco Bay. The Achumawi word for `dog' optionally has a diminutive suffix (caHo'mak!a, `little slave/captive/subordinate one'), and there is another word ?a?la'?mugi? that means `dog' but not `horse' or `slave'. Before you get too excited, let me tell you that this appears to be a descriptive term for a dog whose ears hang down; similarly, Yana cahtumal?gu `dog', lit. `hang-ears'. In an Achumawi Prometheus myth, such a dog brings back fire concealed in his ear. A cognate term `dog-ear' is used for a basketry design, so it is well embedded in the culture. Utterly no basis for a taxonomy associating dogs with e.g. foxes, wolves, or coyotes, or them with one another. References, please. What were the claims, exactly? What was the claimed basis for them? Was the investigator comparing native taxonomies or translations thereof into English? With virtual certainty, the latter. Is your reference to original sources in the anthropological literature or to secondary or tertiary sources there, or to n-ary sources in the philosophical literature? This sort of philosophy strikes me as systematized ethnocentrism. Go ahead and claim that the world must be thus and so because every reasonable person you know sees it that way. But don't go dragging the Indians into it. God knows, they've suffered indignities enough! Bruce Nevin bn@cch.bbn.com (This is my own personal communication, and in no way expresses or implies anything about the opinions of my employer, its clients, etc.) ------------------------------ End of AIList Digest ******************** 5-Oct-87 00:05:18-PDT,16248;000000000000 Mail-From: LAWS created at 5-Oct-87 00:02:23 Date: Mon 5 Oct 1987 00:00-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #229 - Seminar, Conferences To: AIList@SRI.COM AIList Digest Monday, 5 Oct 1987 Volume 5 : Issue 229 Today's Topics: Seminar - Constraint-Posting Planning (BBN) Conferences - Fall Joint Computer Conference 87 & 3rd Applications of AI in Engineering & ASME Computers in Engineering ---------------------------------------------------------------------- Date: Wed 30 Sep 87 14:11:26-EDT From: Marc Vilain Subject: Seminar - Constraint-Posting Planning (BBN) BBN Science Development Program AI/Education Seminar Series Dominance and Subsumption in Constraint-Posting Planning Michael Wellman MIT Artificial Intelligence Lab (MPW@ZERMATT.LCS.MIT.EDU) BBN Laboratories Inc. 10 Moulton Street Large Conference Room, 2nd Floor 10:30 a.m., Friday, October 2nd 1987 Abstract: By integrating a dominance prover into the plan search process, the traditional constraint-posting planning paradigm can be generalized to permit partially satisfiable goals. In this approach, the view of planning as theorem proving is retained, although the emphasis is on deriving facts about the admissibility of classes of candidate plans. Plan classes are generated by posting constraints at various levels of abstraction, then classified within a plan specialization graph that manages inheritance of properties and dominance characteristics. Efficient computation of plan class subsumption is essential for effective use of dominance results. I illustrate this planning framework with examples from SUDO-Planner, an application to medical therapy currently under implementation. Medical therapy has been an unattractive domain for AI planning techniques because of the omnipresence of uncertainty and partially satisfiable objectives. SUDO-Planner's knowledge base contains descriptions of therapy actions at multiple levels of abstraction, with effects represented by qualitative probabilistic influences. The nature of the dominance results derivable by SUDO-Planner suggest that many "metaplanning" rules may be recast as dominance conditions at sufficiently high levels of abstraction. ------------------------------ Date: Fri, 2 Oct 1987 15:28 CST From: Leff (Southern Methodist University) Subject: Conference - Fall Joint Computer Conference 87 List of AI related Papers in Fall Joint Computer Conference 87 October 25-29, 1987, Infomart Tuesday, October 27 10AM - Noon Expresss: Rapid Prototyping and Product Development Via Integrated, Knowledge-Based, Executable Specifications P. Topping, J. McInroy, Lockheed; W. M. Lively, S. Sheppard, Texas A&M University Tuesday October 27 2-3:30 Deriving Contingencies Among Diagnostic Tests with Prolog by Code Examination R. Denney, Schlumberger Well Systems A concuncurrent Multi-Paradigm List Processor TAO/ELis I. Takeuchi, H. G. Okuno, N. Ohsato, M. Kamino, K. Yamazaki, NT&T, Japan Wednesday, October 28 10AM-Noon An Approach to Integrating Expert System Components into Production Sotware W. B. Frakes, C. J. Fox AT&T Bell Labs A Variation of Conceptual Graphs: An Object Oriented Approach T. R. Hines, E. A. Unger, Kansas State University Software Reusability and Knowledge Engineering M. M. Tanik, D. Y. Y. Yun, W. Yin, Southern Methodist University T. J. Lee, A. G. Dale, the University of Texas at Austin A Parallel Algorithm fo r Execution of Production Systems on HMESH S. B. Tien, C. S. Raghavendra, University of Southern California Wednesday, October 28 2-3:30 PM The Intelligent Machines Project in China Chengwei Wong, Beijing Institute of System Engineering, China DFM: The Dataflow Machine for Highly Parallel Symbol Manipulation K. Amamiya, M. Takesue, R. Gasegawa, H. Mikama, NT&T, Japan Wednesday, October 28 3:45-5:15 PM New Methods fo rReal-Time and Image Recognition O. K. Ersoy, D. Y. Kim, Purdue University Dynamic Elastic Interpretation for 3D Objects Reconstruction from Serial X-Sectional Images W. C. Lin, C. C. Liang, Northwestern University; C. T. Chen, University of Chicago Representation and Recognition of Objects From Depth Maps J. K. Aggarwal, B. C. Vemuri, The University of Texas at Austin Thursday, October 29 10:30AM - NOON Frame Synthesis and Inheritance Systems M. Kim, A. S. Maida, Pennsylvania State University A Knowledge-Based Message Generation System for the Nonverbal Profoundly Motor Disabled B. K. Sy, J. R. Deller, Jr. Northeastern University A Uniform Architecture for Rule-Based Meta Reasoning and Representation A. S. Maida, Pennsylvania State University Thursday, October 29 2-3:30 PM A Knowledge-Based Approach to Multiple Query Processing in Distributed Database Systems T. J. Teorey, J. T. Park, The University of Michigan Parallel Execution of Logic Programs in the Framework of OR-Forest Y. G. Tzu, Changsha Institute of Technology, China A Dietary Recommendation Expert System Using OPS5 C. Kao, C. J. Hwang, Purdue University A Conceptual Model for Case Grammar Analysis K. Efe, P. A. Ng, New Jersey Institute of Technology An Analysis of the Knowledge Used for a Structured Selection Problem P. K. Fin,k, F. A. Iddings, M. A. Overby, Southwest Research Institute An Architecture for Adaptive Learningin Rule-Based Diagnostic Expert Systems D. C. St. Clair, W. E. Bond, B. B. Flachsbart, A. G. Vigland, McDonnell Aircraft Corporation Wednesday, October 28 6-7:30PM The Development of Expert Systems - Some Pragmatic Issues Ms. Lorraine M. Duvall, Duvall Computer Technologies, Panel Chair N. J. Martin, SoftPert Systems Inc.; C. J. Green Structured Systems and Software Inc. Thursday, October 29 3:45-5:15PM A Knowledge-Based Aproach to Computer-Aided Design of Structures H. Adeli, K. V. Balasubramanyam, Ohio State University 2 Piece Jig-Saw Puzzle Robot Assembly with Vision, Position and Force Feedback G. C. Burdea, New York Proteus-1: A General Accelerator for CAD S. P. Smith, B. Wood, J. Little, P. Hunter, MCC (Also Panel Discussion on AI and Software Engineering with Dr. David Yun, Southernm Methodist University as Panel Chair and R. Balzer, Information Sciences Institute, C. V. Ramamorthy, University of California at Berkeley, W. W. Royce, Lockheed Software Technology Center, M. M. Tanik, Southern Methodist University, W. Bledsoe, MCC, Roger Bates, Texas Instruments) ------------------------------ Date: Tue, 29 Sep 87 23:06:01 EDT From: sriram@ATHENA.MIT.EDU Subject: Conference - 3rd Applications of AI in Engineering FIRST CALL FOR PAPERS THIRD INTERNATIONAL CONFERENCE ON APPLICATIONS OF ARTIFICIAL INTELLIGENCE IN ENGINEERING AUGUST 8TH-12TH, 1988 STANFORD, CALIFORNIA, USA INTRODUCTION The Third International Conference on AI in Engineering will be held during the second week of August, 1988 in Stanford, California. The first and second international conferences stimulated significant presentations on both the tools and techniques required for the successful use of AI in engineering and many new applications. The organising committee members anticipate that the third conference will be even more successful and encourage papers which describe recent work to be submitted. OBJECTIVES The purpose of this conference is to provide an international forum for the presentation of work on the state-of-the-art in the applications of artificial intelligence to engineering problems. It also aims to encourage and enhance the development of this most important area of research. CONFERENCE THEMES The following application areas and topics are suggested and other related areas will be considered: -------------------------------------------------------------- | | | Application Areas Topics | | Design Representation | | Diagnosis/Evaluation Problem Solving | | Process Control and Planning Constraint Reasoning | | Robotics Learning | | Tutoring Qualitative Models | | Sensing and Interpretation Tools | | User Interfaces | | | -------------------------------------------------------------- INVITED SPEAKERS Keynote: Dr Raj Reddy, Director of the Robotics Institute, Carnegie Mellon University and President, AAAI. Invited Speakers: Dr. Rick Hayes-Roth, Chief Scientist, Teknowledge Others will be announced shortly. SUBMISSION REQUIREMENTS Authors are invited to submit full papers, preferably not exceeding 8,000 words.Each paper should include an abstract and have sufficient details, such as the type of knowledge representation, problem solving strategies, and the implementation language used, to permit evaluation by a committee consisting of renowned experts in the field. Each paper should be accompanied by the following details on the first page: author's name, address, affiliation, the name and address, e-mail, telex and fax of the person to whom all correspondence should be sent, and an indication of the application area and the topic(s). To allow for blind refereeing, the second page should commence with the paper title and abstract and not include any identifying material from the first page. Final instructions on typing format will be forwarded to authors of each accepted paper after refereeing. Four copies of the paper should be submitted to: Professor John Gero Technical Chair, AIE88 Department of Architectural Science The University of Sydney NSW 2006 Australia Tel: 61-2-692-2328 (International) Tlx AA26169 Fax: 61-2-692-3031 Net Address: ARPA: john%archsci.su.oz@uunet.uu.net UUCP: uunet!munnari!archsci.su.oz!john CSNet: john@archsci.su.oz In addition, one copy of the paper should be sent to: Dr R. Adey Computational Mechanics Institute 25 Bridge Street Billerica, MA 01821 Tel. No: 617-667-7582 All papers should be submitted before January 15, 1988. Notification of acceptance will be sent before March 15, 1988. Final copies of the papers are due on or before April 8, 1988. REVIEW CRITERIA All papers will be reviewed by at least two experts in the area. Acceptance of the paper will be based on the quality of the work and its presentation in the paper. ORGANIZING COMMITTEE General Chair Dr. R. Adey, Computational Mechanics Institute, USA Technical Chair Prof. J. Gero, University of Sydney, Australia ADVISORY BOARD Consists of renowned researchers in the field. ------------------------------ Date: Tue, 29 Sep 87 10:28:07 EDT From: decvax!cvbnet!cheetah!rverrill@decwrl.dec.com (Ralph Verrilli) Subject: Conference - ASME Computers in Engineering CALL FOR PAPERS 1988 ASME INTERNATIONAL COMPUTERS IN ENGINEERING CONFERENCE AND EXHIBITION SAN FRANCISCO HILTON SAN FRANCISCO, CALIFORNIA July 31 - August 3, 1988 REAL WORLD APPLICATIONS OF EXPERT SYSTEMS AND ARTIFICIAL INTELLIGENCE The theme for the 1988 ASME International Computers in Engineering Conference will focus on the emerging applications of expert systems and artificial intelligence. This conference and exhibition provides a forum for engineers, managers, researchers, vendors, and users to discuss relevant issues, and to present ideas on computer technology and its impact on the engineering workplace. Over 80 papers and panel sessions are planned covering a broad spectrum of technical computing and computers in the engineering community. The topics covered will encompass: computer aided design and manufacturing, computer simulation, robotics, interactive graphics, finite element techniques, microprocessors, computers in educations, expert systems, and artificial intelligence. Papers are solicited in all areas related to the application, development, research, and education with computers in mechanical engineering. Contributions in the form of full-length papers or extended abstracts are solicited. Accepted papers will be published in the bound Conference Proceedings. Full length papers of special note will be reviewed after the conference for publication in the Society's magazine "Computers in Mechanical Engineering (CIME)". The annual event is sponsored by the Computers in Engineering Division of the American Society of Mechanical Engineers (ASME). San Francisco is the site of this years conference. DEADLINES : Submission of three copies of draft contributions (paper or extended abstract) November 30, 1987 Notification of acceptance to authors February 15, 1988 Submission of author-prepared mats April 1, 1988 For the following technical areas please send papers to the respective program chairmen : { Computer Aided Manufacturing, Computer Simulation, Turnkey CAD/CAM, Integration of CAD and CAM, Computer Aided Testing, Computer Aided Design, Interactive Graphics : Dr. Donald Riley Dept. of Mechanical Engineering University of Minnesota 111 Church Street Minneapolis, MN 55455 612-625-0591/1809 } { Artificial Intelligence, Knowledge Based Systems : Mr. M.F. Kinoglu AI and Expert Systems Group Control Data Corporation 1450 Energy Park Drive Saint Paul, MN 55108 612-642-3817 } { Microprocessors, Robotics, Special Purpose Computers, Man-Machine Interfaces : Mr. David W. Bennett Battelle Pacific Northwest Labs P.O. Box 999 Richland, WA 99352 509-375-2159 } { Robotics in Education, Teaching CAD in Higher Education, University - Industry Collaboration, Microcomputers in the Classroom, Computer-Aided Learning : Dr. Gary Kinzel Ohio State University Dept. of Mechanical Engineering 206 West 18th Street Columbus, Ohio 43210 614-292-6884 } { Finite Element Techniques, Software Standards, Computational Geometry : Dr. Kumar K. Tamma Dept of Mechanical Engineering and Aerospace Engineering West Virginia University Morgantown, West Virginia 304-293-4111 } { Computers in Energy Systems, Computational Fluid Dynamics, Computational Heat Transfer, Combustion Modelling, Process Control : Dr. Ahmed A. Busaina Dept. of Mechanical Engineering Clarkson University Potsdam, New York 315-268-6574 } Topics not in the above categories contact Technical Program Chairman : Mr. Edward M. Patton US Army Ballistic Research Lab Aberdeen Proving Grounds, MD 21005 301-278-6805 ------------------------------ End of AIList Digest ******************** 7-Oct-87 23:14:51-PDT,19569;000000000000 Mail-From: LAWS created at 7-Oct-87 23:11:16 Date: Wed 7 Oct 1987 23:08-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #230 - Speech Databases, Temporal Representation To: AIList@SRI.COM AIList Digest Thursday, 8 Oct 1987 Volume 5 : Issue 230 Today's Topics: Queries - Structural Design and Fabrication & 500K Interconnections Per Second, Literature - Annual Review of Computer Science, Databases - Speech Databases, Representation - Time, Bibliography - Temporal Representation and Reasoning ---------------------------------------------------------------------- Date: Thu, 1 Oct 87 01:39 EDT From: UKJWERK@VAX1.CC.LEHIGH.EDU Subject: Project - Structural Design and Fabrication From: "Keith J. Werkman (KJW1 @ LEHIGH) ATLSS Project B1" I would like to be added to your mailing list. I am interested in Expert Systems related to the Engineering field....Civil Engineering to be exact. My PhD research area is in the Design and Fabrication of Steel Structures. The project, sponsored by NSF, is part of Lehigh University's Engineering Research Center called ATLSS for Advanced Technology in Large Structural Systems. My assigned project is called the Designer/Fabricator Interface or interpreter (DFI). We are looking into developing a system to help structural steel designers better understand the steel fabricator's points of view and, at the same time, allow the fabricator better understand how/why the designer has ordered the specific fabrication in his drawings. Thus, the system needs to present points of view. Current approach includes a frame based system in Quintus Prolog on a Sun 3/160C. The system includes such things as graphics (sunGKS) for input and display of shop drawings in an effort to better describe events as preceived by various parties involved. Hopefully, the result will be a system that will help the mom-and-pop design and fabrication firms throughout our country, hence (and this is the NSF main point) make the US construction industry more competative in the US marketplace. Thus, an interesting topic area with some real potential goals. If you know of any net messages/listing/notices/wispers of ES areas any AI/ES related to CE, structural design, fabrication, welding, erection, connection design, I would be most interested. Since I am the only CS grad student currently involved with network SIGS in the ATLSS center, I feel that any info I can get on current ES in CE might be useful to others research projects such as connection design , material research, business related topics to ES in CE, etc... Keith J Werkman Lehigh University CSEE Department Packard Laboratory, #19 Bethlehem, PA 18015 (215)785-4508 BITNET: KJW1%LEHIGH.BITNET@WISCVM.WISC.EDU Nets: ihnp4!c11ux!lehi3b15!scarecrow!keithw kjw1@lehigh.BITNET keith@lehigh.EDU ------------------------------ Date: 5 Oct 87 06:17:13 GMT From: munnari!mulga.oz!jayen@uunet.UU.NET (Jayen Vaghani) Subject: 500K interconnections - what does that mean? In the Spang Robinson report, there was mention of neural chips handling 500K interconnections per second? What does this actually mean? How do these chips actually work (basically)? Thanks, Jayen. ------------------------------ Date: 4 Oct 87 23:47:54 GMT From: Walter Maner Subject: Re: Annual Review of Computer Science > > The book "Logical Foundations of Artificial Intelligence" > by M. R. Genesereth, N. J. Nilsson contains the following reference, > Levesque, H., "Knowledge Representation and Reasoning", > Annual Review of Computer Science, 1986 > Can anyone give me information about the above journal? I cannot find it > anywhere. This is not a journal but a book published each year by Annual Reviews, Inc., 4139 El Camino Way, Post Office Box 10139, Palo Alto, CA 94303. The ISBN # of the volume you want is 0-8243-3201-6. The same nonprofit company publishes annual reviews of twenty-six other sciences. The 1986 review of computer science was their first volume in this new series. Your article begins on page 255. -- CSNet : maner@research1.bgsu.edu | CS Dept 419/372-2337 UUCP : {cbatt,cbosgd}!osu-cis!bgsuvax!maner | BGSU Generic : maner%research1.bgsu.edu@relay.cs.net | Bowling Green, OH 43403 Opinion : If you are married, you deserve a MARRIAGE ENCOUNTER weekend! ------------------------------ Date: Mon, 05 Oct 87 10:28:58 -0400 From: "Steven J. Nowlan" Subject: Speech Databases A while back I posted a request for information on publically available speech databases. A number of people sent me requests for this information so I am posting a summary. The best source of these databases in North America is the National Bureau of Standards (NBS). The person to speak to is David Pallett whose phone number is (301) 975-2935. However he is extremely busy. The NBS maintains copies of several speech databases for isolated word or connected digit recognition. They make copies of these databases available on various media so that interested parties can copy the database for their own uses. However, the waiting lists for most of these databases are very long! Here is a brief summary of what is available: 1. TI Isolated Word Database - digits, 10 control words, alpha-set Multi-speaker, isolated word 2. VERBEX Database - eleven digits, multi-speaker, isolated word 3. FAA Database - 68 word vocab., multi-speaker, isolated word, phone lines 4. TI Connected Digits Database - variable length digit strings, multi-speaker multi-dialect Access to the above is obtained by contacting David Pallett. 5. AFTI/F-16 - 70 word vocab., multi-speaker, high-noise Contact: Dr. Thomas J. Moore, Biological Acoustics Branch Air Force Aerospace Medical Research Lab, Wright-Patterson AFB, OH 45433. 6. MIT ICE CREAM database - connected sentences (1000 different sentences) multi-speaker. Contact: David Pallett (Available end 87) 7. DARPA "Spelling Bee" Database - sentences of form "word spelling", 600 word vocab., multi-speaker, Contact: David Pallett (Avail 87?) There are also a couple of DARPA databases available to the DARPA contractor community, which are not yet public access, but may be in the near future. Thanks to everyone who provided me with information, and I hope others may find the above information useful. Steve Nowlan Arpanet: nowlan%ai.toronto.edu@relay.cs.net CSNet,Bitnet: nowlan@ai.toronto.edu EAN,X.400: nowlan@ai.toronto.cdn UUCP: {uunet,watmath}!ai.toronto.edu!nowlan ------------------------------ Date: Mon, 5 Oct 87 12:08:38 EDT From: rapaport@cs.Buffalo.EDU (William J. Rapaport) Subject: representation of time Another representation of time, inspired by Allen's work, but differing in significant ways, is in: Michael J. Almeida, "Reasoning about the Temporal Structure of Narratives," Tech. Report 87-10 (Buffalo: SUNY Buffalo Dept. of Computeer Science, 1987). Copies may be had by contacting: library@cs.buffalo.edu William J. Rapaport Assistant Professor Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3180 uucp: ..!{ames,boulder,decvax,rutgers}!sunybcs!rapaport csnet: rapaport@buffalo.csnet internet: rapaport@cs.buffalo.edu [if that fails, try: rapaport%cs.buffalo.edu@relay.cs.net or: rapaport%cs.buffalo.edu@csnet-relay. ] bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: Fri, 2 Oct 87 17:03:00 PDT From: rshu@ADS.ARPA (Richard Shu) Subject: Bibliography on Temporal Representation and Reasoning Ken, I've been reading up on temporal representation and reasoning recently and have compiled the following bibliography. Please distribute it to ailist if appropriate. Richard Shu @TechReport{allen81a, key"allen81a", author "ALLEN, J.F.", title "Maintaining Knowledge About Temporal Intervals, TR 86", institution "University of Rochester, Department of Computer Science", year "1981"} @TechReport{allen81b, key"allen81b", author "ALLEN, J.F.", title "A general model of action and time, TR 97", institution "University of Rochester, Department of Computer Science", year "1981"} @InProceedings{allen81c, key "allen81c", author "ALLEN, J.F.", title "An Interval-Based Representation of Temporal Knowledge", booktitle "Proceedings of 7th IJCAI", organization "IJCAI", pages"221-226", month "August", year "1981"} @Article{allen83a, key "allen83a", author "ALLEN, J.F.", title "Maintaining Knowledge About Temporal Intervals", journal "Communications of the ACM", volume "26(11)", pages "832-843", year "1983"} @InProceedings{allen83b, key "allen83b", author "ALLEN, J.F. & KOOMEN, J.A.", title "Planning using a temporal world model", booktitle "Proceedings of 8th IJCAI 1983", organization "IJCAI", year "1983"} @Article{allen84, key "allen84", author "ALLEN, J.F.", title "Towards a general theory of action and time", journal "Artificial Intelligence", volume "23(2)", pages "123-154", year "1984"} @TechReport{allen85a, key"allen85a", author "ALLEN, J.F. and HAYES, P.J.", title "A Commonsense Theory of Time: The Longer Paper", institution "University of Rochester, Department of Computer Science", year "1985"} @InProceedings{allen85b, key "allen85b", author "ALLEN, J.F. and HAYES, P.J.", title "A Commonsense Theory of Time", booktitle "Proceedings of IJCAI 1985", organization "IJCAI", pages"528-531", year "1985"} @Article{bruce72, key "bruce72", author "BRUCE, B.", title "A Model for Temporal References and its Application in a Question Answering Program", journal "Artificial Intelligence", volume "4", pages "1-25", year "1972"} @TechReport{cheeseman83, key"cheeseman83", author "Cheeseman, P.", title "A Representation of Time for Planning, Technical Note 278", institution "SRI Artificial Intelligence Center", year "1983"} @InProceedings{cheeseman84, key "cheeseman84", author "Cheeseman, P.", title "A Representation of Time for Automatic Planning", booktitle "Proceedings of IEEE International Conference on Robotics", organization "IEEE", year "1984"} @InProceedings{chun86, key "chun86", author "Chun Hon Wai", title "A Representation for Temporal Sequence and Duration in Massively Parallel Networks: Exploiting Link Interactions", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", pages"372-376", year "1986"} @TechReport{dean83, key"dean83", author "DEAN, T.L.", title "Time Map Maintenance", institution "Yale University Computer Science Department", year "1983"} @InProceedings{dean84a, key"dean84a", author "DEAN, T.L.", title "Planning and Temporal Reasoning Under Uncertainty", booktitle "Proceedings of IEEE Workshop on Principles of Knowledge-Based Systems", organization "IEEE", month "December", year "1984"} @InProceedings{dean84b, key"dean84b", author "DEAN, T.L.", title "Managing Time Maps", booktitle "Proceedings of CSCSI 84. Canadian Society for Computational Studies of Intelligence", organization "CSCSI", year "1984"} @TechReport{dean84c, key"dean84c", author "DEAN, T.L.", title "A TNMS User's Manual", institution "Yale University Computer Science Department", year "1984"} @TechReport{dean85, key"dean85", author "DEAN, T.L.", title"Temporal imagery: an approach to reasoning about time for planning and problem solving", institution "Yale University Computer Science Department", year "1985"} @InProceedings{dean85, key "dean85", author "DEAN, T.L.", title "Temporal Reasoning Involving Counterfactuals and Disjunctions", booktitle "Proceedings of 9th IJCAI 1985", organization "IJCAI", pages "1060-1062", month "August", year "1985"} @InProceedings{dean86, key "dean86", author "DEAN, T.L.", title "Intractability and time dependent planning", booktitle "Proceedings of the Workshop on Planning and Reasoning About Action", organization "AAAI", month "July", pages"143-164", year "1986"} @TechReport{fagan80, key "fagan80", author "FAGAN, J.J.", title "Representing Time-Dependent Relations in a Medical Setting", type "Ph.D. thesis", institution "Stanford University", year "1980"} @TechReport{Hanks85, key"Hanks85", author "HANKS, S. and MCDERMOTT, D.", title "Temporal Reasoning and Default Logics", institution "Yale University Department of Computer Science", year "1985"} @Article{hendrix73, key "hendrix73", author "HENDRIX, G.G", title "Modeling Simultaneous Actions and Continuous Processes", journal "Artificial Intelligence", volume "4", pages "145-180", year "1973"} @InProceedings{hirschman81, key "hirschman81", author "HIRSCHMAN, L.", title "Representing implicit and explicit time relations in narrative", booktitle "Proceedings of 7th IJCAI", organization "IJCAI", pages "289-295", month "August", year "1981"} @Article{kahn77, key "kahn77", author "KAHN, K. and GORRY, G.A.", title "Mechanizing Temporal Knowledge", journal "Artificial Intelligence", volume "9", pages "87-108", year "1977"} @InProceedings{kandrashina83, key "kandrashina83", author "Kandrashina, E.Y.", title "Representation of Temporal Knowledge", booktitle "Proceedings of 8th IJCAI 1983", organization "IJCAI", pages "343-345", year "1983"} @InProceedings{ladkin85, key "ladkin85", author "LADKIN, P.B.", title "Comments on the Representation of Time", booktitle "Proceedings of the 1985 Distributed Artificial Intelligence Workshop", year "1985"} @InProceedings{ladkin86a, key "ladkin86a", author "LADKIN, P.", title "Primitives and Units for Time Specification", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "July", pages"354-359", year "1986"} @InProceedings{ladkin86b, key "ladkin86b", author "LADKIN, P.", title "Time Representation: A Taxonomy of Interval Relations", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", pages"360-366", year "1986"} @InProceedings{leban86, key "leban86", author "LEBAN, B., MCDONALD, D and FORSTER, D.", title "A Representation for Collections of Temporal Intervals", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", pages"367-371", year "1986"} @InProceedings{long83b, key "long83b", author "LONG, W. and RUSS, T.", title "A Control Structure for Time Dependent Reasoning", booktitle "Proceedings of 8th IJCAI 1983", pages "230-232", organization "IJCAI", month "August", year "1983"} @InProceedings{malik83, key "malik83", author "MALIK, J. and Binford T.O.", title "Reasoning in Time and Space", booktitle "Proceedings of 8th IJCAI 1983", organization "IJCAI", pages "343-345", year "1983"} @InProceedings{masui83b, key "masui83b", author "MASUI, S., MCDERMOTT, J. and SOBEL, A.", title "Decision-Making in Time-Critical Situations", booktitle "Proceedings of 8th IJCAI 1983", pages "233-235", organization "IJCAI", month "August", year "1983"} @InProceedings{miller85, key "miller85", author "Miller, D., Firby, J. and Dean, T.", title "Deadlines, Travel Time and Robot Problem Solving", booktitle "Proceedings of 9th IJCAI 1985", organization "IJCAI", pages "1052-1054", year "1985"} @Article{mourelatos78, key "mourelatos78", author "MOURELATOS, A.P.D.", title "Events, processes and states", journal "Linguistics and Philosophy", volume "2", pages "415-434", year "1978"} @TechReport{McDermott81, key"McDermott81", author "MCDERMOTT, D.", title "A Temporal Logic for Reasoning about Processes and Plans", institution "Yale University Department of Computer Science", year "1981"} @Article{mcdermott82, key "mcdermott82", author "MCDERMOTT, D.", title "A Temporal Logic for Reasoning about Processes and Plans", journal "Cognitive Science", volume "6", pages "101-155", year "1982"} @TechReport{Moore80, key "Moore80", author "Moore, R.", title "Reasoning about knowledge and action (Technical Report 191)", institution "SRI AI Center", year "1980"} @Book(rescher, key "rescher", author "RESCHER, N.", title "Temporal Logic", publisher "Springer-Verlag", address "New York", year "1971" ) @InProceedings{rit86, key "rit86", author "Rit, J.", title "Propagating temporal constraints for scheduling", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", pages "383-388", year "1986"} @TechReport{shoham86, key "shoham86", author "SHOHAM, Y.", title "Reasoning About Change: Time and Causation from the Standpoint of Artificial Intelligens", type "Ph.D. thesis", year "1986"} @InProceedings{shoham86, key "shoham86", author "Shoham, Y.", title "Reified Temporal Logics: Semantical and Ontological Considerations", booktitle "Proceedings of 7th ECAI, Brighton, U.K.", organization "ECAI", month "July", year "1986"} @InProceedings{shoham86, key "shoham86", author "Shoham, Y.", title "Chronological Ignorance: Time, Nonmonotonicity, Necessity and Causal Theories", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", pages "389-393", year "1986"} @TechReport{Smith83, key "Smith83", author "Smith, S.F.", title "Exploiting Temporal Knowledge to Organize Constraints, Technical Report CMJU-RI-TR-83-12", institution "CMU Robotics Institute", year "1983"} @TechReport{Vere81, key "Vere81", author "Vere, S.A.", title "Planning in Time: Windows and Durations for Activities and Goals", institution "California Institute of Technology Jet Propulsion Laboratory", year "1981"} @TechReport{Vere84, key "Vere84", author "Vere, S.A.", title "Temporal Scope of Assertions and Window Cutoff", institution "California Institute of Technology Jet Propulsion Laboratory", year "1984"} @InProceedings{vere85, key "vere85", author "Vere, S.", title "Temporal Scope of Assertions and Window Cutoff", booktitle "Proceedings of 9th IJCAI 1985", organization "IJCAI", pages "1055-1059", year "1985"} @InProceedings{vilain82, key "vilain82", author "VILAIN, M.", title "A System for Reasoning About Time", booktitle "Proceedings of AAAI-82, Pittsburgh, Pa.", organization "AAAI", month "August", year "1982"} @InProceedings{vilain86, key "vilain86", author "VILAIN, M. and KAUTZ, H.", title "Constraint Propagation Algorithms for Temporal Reasoning", booktitle "Proceedings of AAAI-86, Philadelphia, Pa.", organization "AAAI", month "August", year "1986"} ------------------------------ End of AIList Digest ******************** 11-Oct-87 22:19:42-PDT,18964;000000000000 Mail-From: LAWS created at 11-Oct-87 22:08:42 Date: Sun 11 Oct 1987 22:04-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #231 - Neural Networks, Common Lisp To: AIList@SRI.COM AIList Digest Monday, 12 Oct 1987 Volume 5 : Issue 231 Today's Topics: Queries - Beginning Text for AI and LISP & Beta Sites for Lucid Common Lisp for the Sun 4 & Planning Knowledge and Representation & Public Domain PROLOG via FTP & Common Loops & Scheme on the SUN & Commercial Uses for Neural Nets & Neural Network Texts, Neural Networks - 500K Connections per Second, AI Tools - CMU Common Lisp Distribution ---------------------------------------------------------------------- Date: 8 Oct 87 09:53:05 GMT From: marshek@ngp.utexas.edu (MAt) Subject: Re: A good beginning text for AI & LISP Hi there, ***PLEASE E-MAIL RESPONSES TO ME*** I would like to know about a good introductory text for AI and the LISP language. I have a background in C, fortran but I presume that does not help much, does it? Thanx in advance MAt ------------------------------ Date: Thu, 8 Oct 87 09:47:47 PDT From: edsel!sears@labrea.stanford.edu (Steve Sears) Subject: Looking for beta-sites to test Lucid Common Lisp for the Sun 4. Lucid Common Lisp for the Sun 4 will soon be here. We are interested in locating sites that can provide good feedback by exercising our Sun 4 beta-test version. If you are interested, or know someone that may be, please reply. You can also reach me by phone at Lucid, (415) 329-8400. Thanks in advance... Steve Sears ------------------------------ Date: 9 Oct 87 02:26:39 GMT From: lukose@aragorn.cm.deakin.OZ (Dickson Lukose) Reply-to: lukose@aragorn.OZ (Dickson Lukose) Subject: Planning Knowledge & Representation Survey G'Day Colleagues, I'm a newcomer to the planning paradigm and AI in general. My research interest is in planning heuristics. I'm currently doing a survey on:- (1) "planning knowledge", (2) "knowledge representation for planning" and (3) "process of planning" used by various types (eg. nonhierarchical, hierarchical, script-based and opportunistic) of planning systems. (1) Is there anyone who have done the above survey or knows someone who have done so? (2) Is anyone aware of any survey publications related to the above mentioned areas? (3) Has anyone got any bibliography related to the above subject areas? I'm most interested in communicating with researchers currently involved in R&D of "planning systems". Any suggestions or pointers to the above request will be much appreciated. If enough interest shown, I will transmit(e-mail) the results of survey. ------------------------------------------------------------------ Dickson Lukose | UUCP: ...!seismo!munnari!aragorn.oz!lukose Div. Comp. & Maths | ....!decvax!mulga!aragorn.oz!lukose Deakin University | Victoria, 3217 | ARPA: munnari!aragorn.oz!lukose@SEISMO.ARPA Australia | ACSNET: lukose@aragorn ------------------------------ Date: 8 Oct 87 17:47:44 GMT From: aplcen!jhunix!ins_akml@mimsy.umd.edu (Katherine Martha Lai) Subject: Public domain PROLOG available via ftp? Can anyone tell me where there is a public domain PROLOG that I could get via ftp? I would be running it on a Sun 3/160 under UNIX. Thanks!! I am posting this for a friend, so it probably would be best to reply to him directly (Marty Hall) at "hall@hopkins-eecs-bravo.arpa". ------------------------------ Date: Fri, 9 Oct 87 14:59 EST From: STREIFF%HARTFORD.BITNET@WISCVM.WISC.EDU Subject: Common Loops Request Hi, Im looking for a copy of Common Loops. Does any one have a copy or know where i can get one? Were running Lucid V1.2 on Sun3/75's. Thank you. +------------------------------+----------------------------------------+ | S. David Streiff | BitNet : STREIFF@HARTFORD.BITNET | | CE-CIM-EE | SlowNet : Box 2590 200 Bloomfield Ave | | Combustion Engineering | West Hartford, CT. 06117 | | Windsor CT. | MaBell : (203) 726-9117 | -------------------------------+----------------------------------------+ DisClaimer : My employer is not responsible for anything that i say. They will even deny my existance if given a chance. ------------------------------ Date: 8 OCT 87 15:08-N From: U00124%HASARA5.BITNET@WISCVM.WISC.EDU Subject: Request: Scheme on the SUN? Hello Scheme users! Within short we hope to have a Sun 3/60 system considting of 4 units backed by two 141 MB disks and a 60 MB tape streamer. The system comes fully furnished: C, Fortran 77, Pascal and Modula-2 (and UNIFY). We are interested in running Lisp, preferably Scheme because we use it in different courses and we like the IBM-PC implementation!. The question is: is there a implementation of Scheme for the SUN? and if so: Where to get it, how much it costs, etc..? Any other info concerning Lisp and Prolog on the Suns will be apreciated. Dr. Oscar Estevez Chairman Study Programme Medical Informatics, Univ. of Amsterdam. ------------------------------ Date: 4 Oct 87 23:59:33 GMT From: imagen!atari!portal!cup.portal.com!barry_night-person_stevens@uc bvax.Berkeley.EDU Subject: commercial uses for neural nets -- have, also need, info I am working on a study of commercial uses for neural nets. So far, these include processing commercial loan applications, insurance underwriting, insurance claims processing, signature verification, face and/or voice identification for security, financial optimization. I am interested in swapping applications with those who know of others. These applications are using both hardware (such as the Hecht-Nielsen ANZA board) and software systems. Please contact me by phone at 619-755-7231 or in writing: Barry A Stevens Applied AI Systems, Inc. PO Box 2747 Del Mar, CA 92014 ------------------------------ Date: 7 Oct 87 09:50:45 GMT From: plx!titn!jordan@sun.com (Jordan Bortz) Subject: Neural Networks - Pointers to good texts? Hello; I'm looking for some good texts and/or articles on neural networks, in English, and preferably focusing on real life algorithms/implementations rather than obscure mathematics. If you know of any, please let me know by mail, and I'll summarize to the net; as this topic seems to be generating more interest. I know some articles were mentioned earlier, but what about others? Jordan -- ============================================================================= Jordan Bortz Higher Level Software 1085 Warfield Ave Piedmont, CA 94611 (415) 268-8948 UUCP: (decvax|ucbvax|ihnp4)!decwrl!sun!plx!titn!jordan ============================================================================= ------------------------------ Date: Thu 8 Oct 87 14:28:06-EDT From: Dave.Touretzky@C.CS.CMU.EDU Subject: 500K connections per second The inner loop (and most expensive part) of neural net simulations computes for all j the net input to unit j, which is the sum for all i of the output of unit i times the weight Wji on the connection from i to j. This is just a multiply and accumulate loop. In fact, if you choose the right data structures, it's a matrix-vector multiplication. So when someone advertises that their "neurocmputer" does 500K connections per second, they mean it does five hundred fetch-multiply-accumulate operations per second. This is a useful performance measure because it is independent of the number of units and connections in the model being simulated. There is, unfortunately, no such thing as a commercially available neurocomputer. Presumably, a neurocomputer would either be a computer made out of neurons, or a computer whose physical structure in some way resembled that of the nervous system. No product available today meets either of those tests. What people are selling today as "neurocomputers" are just regular old computers with some neural net simulation software. For example, Hecht-Nielsen NeuroComputer Corporation, the outfit that's been running those full-page four-color ads in AI Magazine, sells their ANZA "neurocomputer" for $15,000. The ANZA system is an off-the-shelf IBM PC/AT with an add-on board containing a Motorola 68020 with floating point co-processor and 4Meg of memory. For roughly the same price you could buy a Sun III (same 68020 processor) and run Unix and X-windows instead of PC-DOS. In fact, Hecht-Nielsen will be announcing a version of their simulation software for the Sun in the near future. That doesn't make the Sun III a neurocomputer, but then again, neither is the ANZA. The TRW Mark III is also a coprocessor build out of conventional components, but it attaches to a Vax rather than an IBM PC. The Science Applications Corporation Sigma-1 is a high speed number cruncher based on a Harvard architecture (the single processor has separate data and instruction paths); it is not a neurocomputer. Science Applications recently acquired a Connection Machine which they plan to use for really heavy duty simulations. (Connection machines aren't neurocomputers either; they're much more general purpose than that. See the article by Blelloch and Rosenberg in IJCAI-87 for a report on using a CM2 to simulate learning in neural nets.) The TI Oddyssey DSP (Digital Signal Processor) is another board that does fast matrix-vector multiplies. Like the other products I mentioned, it is a conventional architecture, basically a handful of TMS 98020(?) hardware multiplier chips. I have a special fondness for Texas Instruments because even though they do some interesting neural net research, they never use the misleading term "neurocomputer" in their ads for the Oddyssey. Will there ever be real neurocomputers? Perhaps some day: Some people are building VLSI circuits whose structure is based on an abstract description of neural circuitry. For example, a group at BELLCORE led by Joshua Alspector and Robert Allen has designed a 54-unit "Boltzmann Machine" chip. The 54 neurons are physically implemented as separate processors on the chip, and their N*(N-1)/2 weighted connections are also implemented by separate pieces of circuitry, giving a fully parallel implementation. This is terrific work, but it will be quite a while before it has any commercial impact, because it's hard to put a lot of neurons on one chip, and expensive to communicate across multiple chips. It is possible to cram several hundred neurons on a chip if you go for fixed weights (resistors) rather than variable ones, but then the network can't learn. Carver Mead and Mass Silviotti at Caltech have built a "silicon retina" low level vision chip using analog (!) VLSI circuitry. The chip's architecture was inspired by the way real retinas do computation. There is also work on optical implementations of neural networks, using lasers, two-dimensional or volume holograms, and various mirrors and photosensors. Two of the big names in this area are Dmitri Psaltis (Caltech) and Nabil Farhat (Penn). It will probably take longer for this technology to reach the marketplace than for VLSI-based technologies, as it is in a much earlier stage of development. A group at Bell Labs has been growing real neurons on a special substrate with embedded electrodes, so they can have an electronic interface to a living neural circuit. This is a neat way to study how neural circuitry works, but they only deal with a handful of neurons at a time. I doubt whether it will ever be practical to design special-purpose computers from living neurons. A good place to learn more about neuromorphic computer architectures (a more decorous term than "neurocomputer", in my opinion) is the proceedings of neural net conferences. There's the proceedings of the 1986 Snowbird Meeting on Neural Networks for Computing, published by the American Institute of Physics in New York. There's also the IEEE First International Conference on Neural Networks, which was held in San Diego this past June. And there's the IEEE Conference on Neural Information Processing Systems - Natural and Synthetic, which will be held in Denver, at the Sheraton Denver Tech Center, on November 8-12. This conference was originally to take place in Boulder, but registration has been so heavy it had to move to larger quarters at the last minute. The conference chairman is Yaser Abu-Mostafa at Caltech. Sorry, I don't have information on how to order proceedings from the IEEE conferences. Contact the IEEE. -- Dave Touretzky ------------------------------ Date: Mon, 05 Oct 87 11:41:30 PDT From: Vicki L. Gordon Subject: CMU Common Lisp Distribution With DARPA funding, the University of Southern California's Information Sciences Institute (USC/ISI) is serving as a distribution center for public-domain Common Lisp software packages. The first of these packages includes the source files for CMU Common Lisp (formerly known as "Spice Lisp"). The package also includes an unsupported public-domain version of the OPS5 language that is written in Common Lisp. If there is sufficient interest, and if funding allows, we will later expand the distribution program to include additional software packages from other sources. CMU Common Lisp is a full implementation of Common Lisp, developed as part of the Spice project at Carnegie-Mellon University (CMU). This system runs only on the IBM RT PC, and only under CMU's Mach operating system (superficially similar to 4.3bsd Unix, but with a different internal organization). Since the IBM RT version of Mach is not currently supported outside of CMU, it follows that CMU Common Lisp is *NOT* a system that anyone can obtain and run as is. We are making the sources for this system available because a number of groups have found it to be a useful starting point for their own Common Lisp implementations. Typically, it takes a man-year of effort to port this system to a new machine and operating system (more if the target environment is unusual). Individuals and small research groups who want a Common Lisp for their own use are advised to use one of the commercially available products. This source code is in the public domain and, once you have them, there is no restriction on how you may use them. They are made available as a public service, with no warranty of any sort by the authors, CMU, or USC/ISI, and with no promise of future support. CMU Common Lisp has been heavily used at CMU, but it has not been extensively tested in any systematic way. Questions about the distribution procedure may be directed via electronic mail to ACTION@ISI.EDU, or you may call (213) 822-5511. Bug reports and questions about the code itself should also be directed to SCOTT.FAHLMAN@CS.CMU.EDU. This distribution package contains approximately 7 megabytes of ASCII source files. It can be obtained over the Arpanet/Milnet by establishing an FTP connection to VENERA.ISI.EDU and logging in as "anonymous" (any password can be used). For optimal response time, please conduct your file transfer during our non-primetime hours (1800 to 0800 PDT). The source files are kept in the following directories: /common-lisp/implementation/cmu/code (Runtime system and interpreter for CMU Common Lisp on IBM RT PC under Mach) /common-lisp/implementation/cmu/clc (Compiler from Common Lisp to native code for RT/Mach. Mostly written in Common Lisp itself.) /common-lisp/implementation/cmu/hemlock (Sources for the Hemlock text editor. This editor is similar at user level to Emacs. Written in Common Lisp, but contains some Mach-specific system calls and display code. Uses either X windows or standard termcap terminals.) /common-lisp/implementation/cmu/OPS5 (Portable Common Lisp version of the OPS5 production-system language. A quick and dirty port of the public domain version that was developed originally in Franz Lisp. No support whatsoever is provided.) /common-lisp/implementation/cmu/miscops (Low level support routines for the IBM PC RT: garbage collection, generic arithmetic, etc. Totally machine-specific. Provided as an example of what needs to be done in order to port this lisp to another machine.) /common-lisp/implementation/cmu/icode (Lisp functions implementing the interface between Mach and CMU Common Lisp. Very system-specific. Provided only as an example.) /common-lisp/implementation/cmu/lib (Cursor definition file and spelling dictionary used by the Lisp runtime system). If you would like to order a tape, we will first send you a release form which you are required to sign prior to receiving the tape. When you return the signed release, include a check made payable to USC/Information Sciences Institute for $100.00 to cover the production and mailing costs. Please remember to include a complete return address. The default tape format will be tar 1600 bpi, unless otherwise specified. Currently ISI is only prepared to distribute tapes containing the CMU Common Lisp code to individuals or companies within the United States. We are currently negotiating with the Department of Commerce and CMU to obtain authorization to distribute the code to countries outside of the United States; however, we do not expect approval in the immediate future. The following hardcopy documentation produced by CMU is also available to all recipients at a cost of $20.00 per package (payable to USC/ Information Sciences Institute). The package includes: - "Internal Design of Common Lisp on the IBM RT PC" - "CMU Common Lisp User's Manual, Mach/IBM RT PC Edition" - "Hemlock User's Manual" - "Hemlock Command Implementor's Manual" Please send your request for a tape and/or documentation to ACTION@ISI.EDU, or mail it to the following address: USC/Information Sciences Institute Attn: ACTION 4676 Admiralty Way Marina del Rey, CA 90292 (213) 822-1511 ext 289 ------------------------------ End of AIList Digest ******************** 11-Oct-87 22:24:18-PDT,19681;000000000000 Mail-From: LAWS created at 11-Oct-87 22:18:45 Date: Sun 11 Oct 1987 22:15-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #232 - Time, Financing, Othello, Philosophy To: AIList@SRI.COM AIList Digest Monday, 12 Oct 1987 Volume 5 : Issue 232 Today's Topics: Representation - Time Bibliography, Business - Expert Systems Company Financing, Games - International Computer-Othello Match, Philosophy - Goal of AI & Flawed Minds ---------------------------------------------------------------------- Date: Thu, 8 Oct 87 15:23:36 PDT From: ladkin@kestrel.ARPA (Peter Ladkin) Subject: time bibliography ailist readers might like to note that richard shu's bibliography misses halpern and shoham's paper in LICS 1986, pelavin and allen's paper in the Proceedings of the IEEE for October 1986, and also papers published in AAAI-87 and IJCAI-87. Additionally, there is a large literature on points and periods in philosophy since Rescher and Urquhardt, to which one can get pointers from Kuhn's review of van Benthem that I referenced. In a complete list of temporal reasoning, one should also include the huge amount of literature from program semantics, especially the semantics of concurrency. It was interesting to note that Shu's bibliography and mine were almost disjoint ............ peter ladkin ladkin@kestrel.arpa ------------------------------ Date: 9 Oct 87 14:12:15 GMT From: amdcad!sun!sundc!potomac!grover@ames.arpa (Mark D. Grover) Subject: Re: Bibliography on Temporal Representation and Reasoning I had a relevant article in AAAI 82: Grover, M.D. "A Synthetic Approach to Temporal Information Processing", AAAI-82, pg. 91-4. Based on my PhD dissertation, the work addresses the use of Montague Grammar-based temporal representations of complex English verb tense. - MDG - -- --> Our site is in transition. Please use one of the addresses below. <-- Mark D. Grover (grover@ads.ARPA) UUCP: ames!amdcad!sun!sundc!potomac!grover Advanced Decision Systems 1500 Wilson Blvd #512; Arlington, VA 22209 Future address: grover@dc.ads.com "I've been ionized... but I'm okay now." ------------------------------ Date: 6 Oct 87 02:21:35 GMT From: oliveb!intelca!mipos3!omepd!chedley@ames.arpa (CHEDLEY) Subject: Re: Expert Systems Company Financing... In article <7260@dartvax.UUCP> waltervj@dartvax.UUCP (walter jeffries) writes: > >I am not the business end of things but >would appreciate any comments/experiences that people may have with getting >capital (sources, things to be careful of, etc.). Of course, if you want to >invest money as well as advice that would be appreciated too :-). > There are three major sources of money for start-up financing: 1) Money from the owners/starters of the company: In this case it is your and your partners' own savings and personal loans (credit cards, Home equity loans, personal unsecured bank loans,..) MONEY FROM THIS SOURCE IS TYPICALLY INSUFFICIENT TO GET THE BUSINESS ROLLING 2) Venture Capital: This money belongs to funds(*), companies or private individuals who are looking to invest in start-up businesses. In return they require the "ownership" of a portion of the business, along with some other conditions (oversight on the books of the company, a say in management appointments, options on the share of the company if and when it goes public,...etc) MONEY FROM THIS SOURCE IS RELATIVELY AVAILABLE. 3) Govrmt Money (State/Federal) : This is typically an easy conditions loan (low interest rate, long grace period, easy payment schedule..) provided by some state or federal agencies to promote small and start companies. Try to tap this source to the max. And you do not have to be a woman or a member of a minority group to qualify for this cheap source of financing. GOVRNMT MONEY IS THE CHEAPEST SOURCE OF FINANCING START UPS Due to the constraints of the venture capitalist's money, it is advantageous to leverage it as much as possible with the other sources's money. That is, for each dollar from source 1 or 3, get the maximum venture capital you can reach for. (*): There are even a few venture capital mutual funds out there. ..CHEDLEY.. ------------------------------ Date: 7 Oct 87 18:20:12 GMT From: amdahl!oliveb!epimass!epiwrl!shore@ames.arpa (John Shore) Subject: Re: Expert Systems Company Financing... In article <7260@dartvax.UUCP> waltervj@dartvax.UUCP (walter jeffries) writes: > >I am in the process of starting a company to do expert systems developement >in the field of psychiatry.... >...(sources, things to be careful of, etc.).... Things to be careful of? Expert systems and AI. js ------------------------------ Date: 6 Oct 87 01:17:50 GMT From: mcvax!nikhefk!kvr@uunet.uu.net (kees van rijn) Subject: intercontinental computer - othello match INTERCONTINENTAL COMPUTER - OTHELLO MATCH Last saturday there was a computer-othello match of MY TURN in The Netherlands with BILL in the USA. MY TURN has been written by Cas Wilders and won the preceding local mini-tournament in Amsterdam with REV87 (by Joost Buys), MAST87 (by Ron Kroonenberg) and BADIA1.2 (by Marcel van Tien). BILL is vice-champion of the USA since last US' tournament, some years ago. BILL has been written by Kai-Fu Lee and Sanjoy Mahajan. Communication between Pittsburgh USA and Amsterdam NL took place via EARN / BITNET. After this match, there were also games of REV87 and MAST87 with BILL via this communication channel. All games were won by BILL. In the first match MY TURN got low mobility because of a wrong move in the beginning. It was hopeless to continue and Cas resigned. For the other two games, the level of the participants was probably near equal, though initially REV87 had also problems with mobility, but it recovered. For non-experts in othello, like me, it is however very difficult to estimate the real value of a position. All of us agreed that it is a very hard job to improve strength of the programs further with known techniques. According to Kai-Fu, faster machines lead only to marginal improvement, and better search algorithm is too hard. We think that most improvement of last years is from implementation of specific othello knowledge into the programs. However, probably the level of present programs is so high, that in a tournament of best computer programs with best human players, computers will win more than 80% of the games. Technically, the communication channel was good, though exact time checking was impossible because of a delay of about 5 seconds before a move arrived. This time is not garanteed, and it is also not yet possible for the participants to check the time that a message was sent. Another problem was that backspaces from Amsterdam were not executed, but turned into periods, so that careful typing was required. We were later told that delete probably would have been effective. And sometimes, messages from other people were disturbing clear communication. Generally speaking however, the match passed off very successfully. kees van rijn (organizer) ------------------------------ Date: 6 Oct 87 21:04:05 GMT From: PT!speech2.cs.cmu.edu!kfl@cs.rochester.edu (Fu Lee) Subject: Re: intercontinental computer - othello match In article <248@nikhefk.UUCP>, kvr@nikhefk.UUCP (kees van rijn) writes: > > INTERCONTINENTAL COMPUTER - OTHELLO MATCH > > Last saturday there was a computer-othello match of > MY TURN in The Netherlands with > BILL in the USA. > .... Thanks to Kees for organizing this match, and for this accurate commentary. There were, however, a few misunderstandings which I hope to clarify. > All of us agreed that it is a very hard job to improve > strength of the programs further with known techniques. > According to Kai-Fu, faster machines lead only to marginal > improvement, and better search algorithm is too hard. Actually, I believe faster machines will lead to substantial improvement, as they did for chess. However, making othello hardware is not as fruitful since current programs already outplay humans, and since there are no incentives. I think an improved search algorithm is both more effective and intellectually satisfying. However, our attempts in the past year have not been encouraging. >We think that most improvement of last years is from implementation >of specific othello knowledge into the programs. Actually, the two major contributions from BILL are: (1) encoding all Othello knowledge into tables for fast evaluation, and (2) Bayesian learning of how to combine evaluation features. > kees van rijn > (organizer) Kai-Fu Lee ------------------------------ Date: 4 Oct 87 18:23:43 GMT From: ihnp4!homxb!mtuxo!mtune!codas!usfvax2!pdn!alan@ucbvax.Berkeley.E DU (Alan Lovejoy) Subject: Re: Goal of AI: where are we going? In article <46400008@uxe.cso.uiuc.edu> morgan@uxe.cso.uiuc.edu writes: /Maybe you should approach it as a scientist, rather than an engineer. Think /of the physicists: they aren't out to fix the universe, or construct an /imitation; they want to understand it. What AI really ought to be is a /science that studies intelligence, with the goal of understanding it by /rigorous theoretical work, and by empirical study of /systems that appear to have intelligence, whatever that is. The best work /in AI, in my opinion, has this scientific flavor. Then it's up to the /engineers (or society at large) to decide what to do with the knowledge /gained, in terms of constructing practical systems. The word "artificial" implies either an imitation or synthetic object, or the general/abstract laws governing an entire class of such objects. The question is, does "artifical intelligence" mean "synthetic and/or imitation intelligence" (most computer programs currently fall into this category :-) ) or "real intelligence exhibited by artifical systems"? Is AI mostly concerned with the *faking* of intelligence, with intelligence per se or with intelligence as exhibited by artificial systems? Given the current state of the art, perhaps it should be called "Real Stupidity". (Only half :-) ). The "scientific" study of intelligence would involve such subfields as cognition, semantics, linguistics, semiotics, psychology, mathematics, cybernetics and a host of other disciplines I can't think of right now, some of which probably don't exist yet. Creating an intelligent "artifact" (artificial intelligence) is only a "scientific" endeavor to the extent it serves as experimental proof (or refutation) of some *scientific* theory, or else as the raw data from which a theory is induced. If the purpose of AI is to build a computer just as smart as a human being because that would be a useful tool, then it's engineering. If the purpose is to prove or induce theories about intelligence, then it's scientific. It appears that both cases probably apply. It is disturbing how often "science" is confused with "technology" and/or "engineering". People also tend to forget that science involves both the formulating of theories AND experiments. Experiments often require a great deal of mundane (and sometimes not so mundane) engineering work. AI came about because computers opened up a whole new way to experimentally test theories about intelligence. Physicists might very well try to construct an "artificial" universe, if it would help to prove or induce a physical theory (the "Big Bang", for instance). They'd probably require a lot of help from the engineers, though (and probably a permit from the EPA :-) ). --alan@pdn ------------------------------ Date: 5 Oct 87 23:11:11 GMT From: PT!isl1.ri.cmu.edu!cycy@cs.rochester.edu (Christopher Young) Subject: Re: Goal of AI: where are we going? In article <1330@houdi.UUCP>, marty1@houdi.UUCP (M.BRILLIANT) writes: > Point two, we keep using the human mind as a tool, to solve problems. > As such, it is not merely a phenomenon, but a means to an end, and is > subject to judgments of its utility for that purpose. Now we can say > whether it is perfect or flawed. Obviously, it is not perfect, since > we often make mistakes when we use it. Score one for Ware. This is true. However, this is not the only use for the human mind. The human mind is also used to imagine fanciful dreams, to love and hate and otherwise feel emotion, and to make value judgement even when there is no real logical reason for choosing option one over option two. So perhaps it can be flawed in one way, but not in others (since it is difficult to say what is flawed in some of these instances). -- -- Chris. (cycy@isl1.ri.cmu.edu) I know you believe you understand what you think I said, but I am not sure you realise that what you heard is not what I meant. ------------------------------ Date: 5 Oct 87 22:58:34 GMT From: PT!isl1.ri.cmu.edu!cycy@cs.rochester.edu (Christopher Young) Subject: Re: Goal of AI: where are we going? In article <270@uwslh.UUCP>, lishka@uwslh.UUCP (Christopher Lishka) writes: > To me this seems to be one of many problems in A.I.: the assumption > that the human mind can be looked at as a machine, and can be analyzed > as having flaws or not, and subsequently be fixed or not. > > A comment: why don't A.I. "people" use the human mind as a model, for > better or for worse, and not try to label it as "flawed" or "perfect?" I guess I basically agree, though I certainly feel that there are some people whose reasoning is either flawed or barely existent, and it is true in fact that physiological parameters can affect thought, and that these parameters can be adjusted in certain ways to cause depression, and to recover from depression (etc). So in that way, one might say that human minds may become flawed, I suppose. On the other hand, since we pretty much define "mind" based on human ones, it's hard to say that they are flawed. If there was something "perfect" (whatever that might be", then it might very well not be a mind. I do believe that there is some mechanism to minds (or perhaps a variety of them). One reason why I am interested in AI (perhaps this is very Cog. Sci. of me, actually) is because I think perhaps it will help elucidate the ways in which the human mind works, and thus increase our understanding of human behaviour. I don't know; perhaps I am naive in that respect. At anyrate, I do try to use the human mind as a model in at least some of what I am doing. Just thought I'd throw in my two cents. -- -- Chris. (cycy@isl1.ri.cmu.edu) I know you believe you understand what you think I said, but I am not sure you realise that what you heard is not what I meant. ------------------------------ Date: 4 Oct 87 20:19:03 GMT From: wcalvin@well.UUCP (William Calvin) Reply-to: wcalvin@well.UUCP (William Calvin) Subject: Re: Goal of AI: where are we going? Making AI a real science suffers from the attitude of many of its founders: they'd rather re-invent the wheel than "cheat" by looking at brain research. While Minsky's SOCIETY OF MIND is very interesting, one gets the impression that he hasn't looked at neurophysiology since the 1950s. Contrast that to Braitenberg's little book VEHICLES (MIT Press 1984), which summarizes a lot of ideas kicking around neurobiology at the simple circuit level. The other thing strikingly missing, besides a working knowledge of neurobiology beyond the Hubel-Wiesel level, is a knowledge of evolutionary biology beyond the "survival of the fittest" level. Emergent properties are a big aspect of complex systems, but one seldom hears much talk about them in AI. William H. Calvin University of Washington NJ-15, Seattle WA 98195 ------------------------------ Date: 6 Oct 87 18:23:31 GMT From: bbn!uwmcsd1!uwm-cs!litow@cs.rochester.edu (Dr. B. Litow) Subject: Re: Goal of AI: where are we going? (Where should we go...) > > The principal difficulty in cognitive science is that it is in its > infancy. I think that psychology is today where physics was in > Newton's time. And a LOT of "narrow minded" theories came and went in > Newton's time. Including Newton's theories. > > Steve Frysinger Newton's primary contribution in Principia is a method. The method has NOT been modified at its core in the elapsed three centuries. It is still at the basis of all western physical science. Newton understood its importance as V.Arnold has pointed out in his book Geometric Methods in the Theory of Ordinary Differential Equations (Springer). The method is very simple to state: pose and solve differential equations for the phenomena. Prior to anything else in western physics there is this method. In this respect all of quantum mechanics is only a conservative (almost in the sense of logic) extension of rational mechanics. Incidentally rational mechanics was not developed explicitly by Newton. It is a product of the Enlightenment researchers,e.g. the Bernoullis and especially Euler. Underlying the method is something nameless which when it is finally investigated (the time is approaching) will be a decisive element in actually showing what is really conveyed by the adjective "western". ------------------------------ Date: 5 Oct 87 15:04:54 GMT From: spf@moss.ATT.COM Reply-to: spf@moss.UUCP (Steve Frysinger) Subject: Re: Goal of AI: where are we going? (Where should we go...) In article <493@vax1.UUCP> czhj@vax1.UUCP (Ted Inoue) writes: }Some of you may remember my postings from last year where I expounded on the }virtues of cognitive psychology. After investigating research in this field }in more detail, I came up very disillusioned. Here is a field of study in }which the soul purpose is to scientifically discover the nature of thought. }Even with some very bright people working on these problems, I found that the }research left me cold. Paper after paper describe isolated phenomena, then go }on to present some absurdly narrow minded theory of how such phenomena could }occur. Perhaps you're right; there is not doubt that the system in question is highly complex and interconnected. However, the same claim can be made about the domain of physics. And (in the west at least) research and progress in physics has been built upon small pieces of the problem, complete with small theories (which usually seemed incredibily naive when disproved). Now another approach to physics is possible (see Kapra's "Tao of Physics"). It would probably not be observational (which I require of any science) but introspective instead. Me? I like both. When I do science, I build up from measurable components, creating and discarding petty theories along the way. When I do zen, it's another matter entirely (no pun intended). The principal difficulty in cognitive science is that it is in its infancy. I think that psychology is today where physics was in Newton's time. And a LOT of "narrow minded" theories came and went in Newton's time. Including Newton's theories. Steve Frysinger ------------------------------ End of AIList Digest ******************** 11-Oct-87 22:32:33-PDT,15683;000000000000 Mail-From: LAWS created at 11-Oct-87 22:27:49 Date: Sun 11 Oct 1987 22:21-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #233 - Philosophy To: AIList@SRI.COM AIList Digest Monday, 12 Oct 1987 Volume 5 : Issue 233 Today's Topics: Philosophy - Goals of AI & Flawed Minds ---------------------------------------------------------------------- Date: 5 Oct 87 15:41:48 GMT From: merlyn@starfire.UUCP (Brian Westley) Subject: Re: (Where should we go...) How to get there? Ted Inoue writes: > In article <46400008@uxe.cso.uiuc.edu> morgan@uxe.cso.uiuc.edu writes: > >What AI really ought to be is a > >science that studies intelligence, with the goal of understanding it by > >rigorous theoretical work, and by empirical study of > >systems that appear to have intelligence, whatever that is. > .... > On the other hand, if we take an educated approach to the problem, and study > 'intelligent' systems, we have a much greater chance of solving the mysteries > of the mind... > > ---Ted Inoue This might work, but I would compare your method (understand the human mind first, then mimic it via computer) to be similar to early heavier-than-air experiments. Birds were the only working model, but we never got off the ground until we stopped building airplanes that flapped their wings. Intelligent computers will probably be as different from the human mind as 747's are from hummingbirds. They will both 'think', but in radically different ways. Of course, I could be wrong, so both methods should be explored. Merlyn LeRoy ------------------------------ Date: 8 Oct 87 10:06:00 EST From: cugini@icst-ecf.arpa Reply-to: Subject: flawed minds I'm sure I'm gonna regret getting into this (stop me before I philosophize again), but here goes. Put me down as a "yes" vote on the question if all (the vast majority of ?) human minds are flawed. First, to clear away some underbrush: of course the truth of the statement is relative to the meaning of the words used. "grass is green" is false if the referent of "grass" is zebras, ho hum. To "play fair", it seems to me we should attempt to take the most plausible interpretation of what is after all a pithy statement, and contend with that. It seems to me that "mind" normally means "that which enables the owner of the mind to think" - eg if a Martian had a glarp instead of a brain, but could still play a mean game of chess, and discuss the NFL strike, etc, we surely would agree s/he had a mind. Since it is an *essential* feature of a mind that it enables one to think (positivistic formulation: mind IS the ability to think), it seems fair to say that to the extent one thinks imperfectly, one's mind is flawed. I am blithely assuming that "correct thinking" implies at least the ability to formulate accurate descriptions of the world, and manipulate them so as to draw correct conclusions. I don't claim that a mind is nothing but an implementation of logic, but it at least ought to be logically correct as far as it goes. Insofar as the human mind implements unsound logic, it is flawed (lots of people, eg, fall into the fallacy of the converse, multiply incorrectly, etc.) "the human mind is flawed" thus seems to me the same kind of statement as "XYZ cars don't work well". Of course, considered qua phenomenon, an XYZ car is neither good, bad, nor ugly. But insofar as one accepts the bland (?) assumption that the essential purpose of a car, qua car, is to transport you from A to B, via roadway, then the question is merely whether XYZ cars in fact usually succeed or not in this task. The essential purpose of a mind, qua mind, is, among many other things, to draw conclusions correctly from a given set of facts. To the extent it fails to do so, it is flawed. John Cugini ------------------------------ Date: Thu 08 Oct 1987 08:33 CDT From: UUCJEFF%ECNCDC.BITNET@WISCVM.WISC.EDU Subject: THE MIND I read some of the MIND theories espoused in the Oct 2 list, and am frankly disappointed. All those debates are based on the Science vs Mysticism debates that were going on 10 years ago when I was an undergrad. I have since discarded both arguments into the /dev/null file. Nonetheless I would like to make a few comments. 1) It is wrong to assume emotion is a flaw of the mind, or even bring up Manson and Hitler. I would say absence of emotion is a flaw of the mind. You want to talk about genius where mind and emotion are equal partners, look at Ornette Coleman or John Coltrane. Anyone who downgrades emotion (or i should say "emotional intelligence") is committing suicide. 2) Even if you say a mind is flawed because it can't be "objective", ( I know some cyberneticians who were saying that we soon won't be talking in terms of "objective" vs "subjective". Those words will be obsolete) let me ask a question. Does anyone believe that as two people become more informed about any subject, as their knowledge and information increases that they will become in agreement? I think the answer is no, and not because the mind is flawed. 3) Some of you seem to be making science in general and AI in particular a religion. Especially with pie-in-the-sky projects of making computers AI identical to human intelligence. That strikes me as another immortality project. Let us say for the sake of argument that you could ( sometime in the year 2525). In that case the product will be necessarily flawed since the human mind is flawed by your arguments. So what have your gained. 4) In the area of art, I prefer so-called irrationality and surrealism. it is more interesting. 5) AI should concern itself with solving problems, discovering new ways to solve and conceptialize problems. It is not as glamorous as making artificial souls, but more practical and fruitfull. Jeff "FREE" Beer, PAN recording artist ------------------------------ Date: 7 Oct 87 14:56:14 GMT From: umn-d-ub!umn-cs!ramarao@rutgers.edu (Bindu Rama Rao) Subject: Re: Goal of AI: where are we going? Is the Human mind flawed? Can we pass such a judgement without knowing anything about the human mind? Do we really understand how the mind works? Aren't we trying to model the mind because we are in awe of all the power the mind posesses? Is the mind flawed just because humans make decisions based on their emotional involvement? Isn't the mind used for analysis only while emotions play a major part in formulating the final decision? Let's not hastily dismiss the human mind as flawed. -bindu rama rao. ------------------------------ Date: 9 Oct 87 15:32:59 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: Goal of AI: where are we going? In article <2281@umn-cs.UUCP>, ramarao@umn-cs.UUCP (Bindu Rama Rao) writes: > > Is the Human mind flawed? > > Can we pass such a judgement without knowing anything about the human mind? > > Do we really understand how the mind works? Let's draw an analogy. You are driving an X-Brand car from Pittsburgh to Atlanta and halfway there it bursts into flame. Without knowing how the car works you can conclude it was flawed. Mr X. goes to an employment interview and gets angry or flustered and says something that causes him to be rejected. Without knowing how his mind works you can conclude it was flawed. > Aren't we trying to model the mind because we are in awe of all the > power the mind posesses? Of course we are. But saying the mind is enormously powerful is not contradicted by saying it's not perfect. A car with a big engine is enormously powerful and almost certainly not perfect. > Is the mind flawed just because humans make decisions based on > their emotional involvement? Isn't the mind used for analysis only > while emotions play a major part in formulating the final decision? Factually, we know the mind is flawed because we observe that it does not do what we expect of it. As a hypothesis, we can test the idea that it is flawed because of the action of what we call emotions. As a further hypothesis, we can also test the idea that emotions motivate all human activity. Personally, I like both those hypotheses. Question of definition here: do we agree that emotion, reason, consciousness, will, etc., are all functions of the mind? > Let's not hastily dismiss the human mind as flawed. Who's dismissing it? I know my car is flawed, but I can't afford to dismiss it. I'm not dismissing my mind either. How could I? :-) M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 10 Oct 87 11:24:28 GMT From: k.cc.purdue.edu!l.cc.purdue.edu!cik@j.cc.purdue.edu (Herman Rubin) Subject: Re: Goal of AI: where are we going? In article <1368@houdi.UUCP>, marty1@houdi.UUCP (M.BRILLIANT) writes: > In article <2281@umn-cs.UUCP>, ramarao@umn-cs.UUCP (Bindu Rama Rao) writes: > > > > Is the Human mind flawed? > > > > Can we pass such a judgement without knowing anything about the human mind? > > > > Do we really understand how the mind works? > The human mind is definitely flawed, very fortunately. I do not see how an intelligent entity can fail to be flawed if it has only the computing power of the universe available. I define intelligence as the ability to deal with a _totally unforeseen situation_. It is easy to give examples in which the amount of information needed to effect a logical decision would require more memory than the size of the universe permits. Therefore, dealing with such a situation _requires_ that such extralogical procedures as intuition, judgment, somewhat instinctive reactions, etc., must be involved. That is not to say that one cannot find out that certain factors are of lesser importance. But the decision that these less important factors can or should be ignored is still a matter of judgment. Therefore, an intelligent treatment of a problem of even moderate complexity requires that nonrational procedures must be used. These cannot be correct; at most we can determine in _some_ cases that they are not too bad. In other cases, we can only hope that we are not too far off. There is no "rational" intelligent entity for moderately difficult problems! -- Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907 Phone: (317)494-6054 hrubin@l.cc.purdue.edu (ARPA or UUCP) or hrubin@purccvm.bitnet ------------------------------ Date: 10 Oct 87 17:01:55 GMT From: udel!montgome@cs.rochester.edu (Kevin Montgomery) Subject: Re: Goal of AI: where are we going? >> In article <2281@umn-cs.UUCP>, ramarao@umn-cs.UUCP (Bindu Rama Rao) writes: >> > Is the Human mind flawed? C'mon guys, lighten up for a sec. Flawed implies a defect from it's design. Therefore, if someone's mind doesn't do what it's designed to do (namely help keep the organism alive, etc), THEN it's flawed (ex: schizos, manics, etc). A "normal" person does NOT have a flawed mind, just an illogical one. What do you expect when the old brain (producing emotions, feelings and the like) is still in the design? So the $64K answer is: no, the mind is not (usually) flawed, but it is illogical. Is having an illogical mind a problem? Hell no! It's what keeps organisms going- drives for self-preservation, procreation, etc. While striving to be logical IS (i feel) a noble aspiration, there's no way to totally shut out something like emotions so deeply ingrained into the mental architecture. (one may even argue that if we were to consider all things logically, then civilization would die out rather quickly, but i'm not gonna touch that one) At any rate, if you want to do some neato cognitive modelling stuff, then you've got to (eventually) incorporate the functions of the old brain (illogic) with the logical processes we normally consider. If you're gonna do some neato expert system stuff involving pure logic, then don't worry about it. `kay? `kay. -- Kevin Desperately-trying-to-get-into-Stanford Montgomery ------------------------------ Date: 11 Oct 87 18:34:10 GMT From: yale!krulwich@husc6.harvard.edu (Bruce Krulwich) Subject: Re: Goal of AI: where are we going? In article <1368@houdi.UUCP> marty1@houdi.UUCP (M.BRILLIANT) writes: >Factually, we know the mind is flawed because we observe that it does >not do what we expect of it. If I expect my car to take me to the moon and it doesn't, is it flawed?? No, rather my expectation of it is wrong. Similarly, we shouldn't say that the mind is flawed until we're sure that our definition of "intelligence" is perfect. > As a hypothesis, we can test the idea >that it is flawed because of the action of what we call emotions. Why do you assume that emotions are a flaw?? Just maybe emotions are at the core of intellegence, and logic is just a side issue. >As >a further hypothesis, we can also test the idea that emotions motivate >all human activity. Personally, I like both those hypotheses. If you think that emotions motivate all human activity, why do you dismiss emotions as a flaw in the mind?? It seems to me that human activity is a lot more "intelligent" than any AI system as of yet. >Question of definition here: do we agree that emotion, reason, >consciousness, will, etc., are all functions of the mind? Yes, and not necessarily "flawed" ones. Bruce Krulwich ARPA: krulwich@yale.arpa Being true heros, or krulwich@cs.yale.edu they lept into action. Bitnet: krulwich@yalecs.bitnet (Bullwinkle) UUCP: {harvard, seismo, ihnp4}!yale!krulwich (Any B-CC'ers out there??) ------------------------------ Date: 12 Oct 87 00:45:58 GMT From: topaz.rutgers.edu!josh@rutgers.edu (J Storrs Hall) Subject: Re: Goal of AI: where are we going? krulwich@yale.ARPA (Bruce Krulwich): If I expect my car to take me to the moon and it doesn't, is it flawed?? If you expect your car to take you to the moon, then I would say your mind *is* flawed... --JoSH :^) ------------------------------ Date: 11 Oct 87 16:55:37 GMT From: psuvax1!vu-vlsi!swatsun!scott@husc6.harvard.edu (Jay Scott) Subject: Is the human mind flawed? Here's how I think about it: "Flawed" I take to mean "not-good in some particular respect." And "good" does not have a fixed, absolute meaning. If you ask, "Is this rock good?" I have to reply, "What for?" It may be good used as a piece of gravel but bad used as a gemstone! So in the same way, you may ask "Is the human mind flawed?" I answer "Depends. Is there something you wanted to use one for?" If you think minds just are, then "flawed" doesn't apply (neither does "perfect"). But if you want to use a mind to, say, do math, you're likely to be annoyed at its tendency to make mistakes--a flaw, for your purposes. -- Your opinion may vary. I can only define words as I use them, not as you may. Jay Scott ...bpa!swatsun!scott ------------------------------ End of AIList Digest ******************** 11-Oct-87 22:40:05-PDT,10662;000000000001 Mail-From: LAWS created at 11-Oct-87 22:35:41 Date: Sun 11 Oct 1987 22:34-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #234 - Seminars, Statistics Conference To: AIList@SRI.COM AIList Digest Monday, 12 Oct 1987 Volume 5 : Issue 234 Today's Topics: Seminar - Systems with Multiple Expertise (BBN) & Very Large Traveling Salesman Problems (Stanford) & AI in Manufacturing (SU) & Pengi: An Implementation of a Theory of Activity (SU) & Persistence, Intention, and Commitment (AT&T), Conference - AI and Statistics at JSM ---------------------------------------------------------------------- Date: Mon, 28 Sep 1987 10:34 EDT From: Marc Vilain Subject: Seminar - Systems with Multiple Expertise (BBN) [Forwarded from the IRList Digest.] BBN Science Development Program AI/Education Seminar Series SYSTEMS WITH MULTIPLE EXPERTISE Alexander Makarovitsch Computer Science Department, West Catholic University (Angiers, France) BBN Laboratories Inc. 10 Moulton Street Large Conference Room, 2nd Floor 10:30 a.m., Tuesday, September 29, 1987 Abstract: By a system with multiple expertise, we mean one containing two or more expert systems, all working on a common information domain. Three such systems under current development will be briefly reviewed. (1) Syclope, a system being developed for the French Board of Electricity for aiding agents working under risky conditions to improve their task behavior; (2) Sonar, a system being developed for the Bull Computer Group for aiding in the design, operation, and maintenance of computer networks; (3) Link, a system being developed for the Bull Group for aiding decision makers in the networked computer area in the processes of product planning. The review will focus on the difficulties encountered in key aspects of the development process: the acquisition of expertise, knowledge representation, man/machine interface, and interactions amopng the system and multiple users and experts. ------------------------------ Date: Fri 2 Oct 87 15:50:59-PDT From: Anil R. Gangolli Subject: Seminar - Very Large Traveling Salesman Problems (Stanford) 15-October-87: David Johnson (AT&T Bell Labs) Near-Optimal Solutions to Very Large Traveling Salesman Problems Most experimental studies of algorithms for the Travel- ing Salesman Problem (TSP) have concentrated on relatively small test cases, instances with 100 cities or less. In practice, however, much larger instances are frequently encountered, both in engineering and scientific applica- tions. This talk begins by surveying complexity results about the TSP and the status of algorithms for finding optimal solutions to small instances. It then goes on to report the results of experiments with standard TSP heuris- tics on large instances, from 500 cities to 100,000, examin- ing the trade-offs obtainable between running time and qual- ity of solution. Most of the standard heuristics will be compared, including such new approaches as ``simulated annealing,'' but particular emphasis will be placed on the acknowledged ``champion,'' the sophisticated Lin-Kernighan algorithm. Using various programming tricks, we have imple- mented a version of this heuristic for the Euclidean case that remains feasible even for 10,000 city instances (8 hours on a minicomputer), and continues to find tours that are within 2% of optimal. For 20,000 or more cities, we could still obtain tours that were within 5% of optimal using Lin-Kernighan as a subroutine in a partitioning scheme suggested by Karp. If one is willing to settle for slightly worse tours, an approximate version of the Christofides heuristic seems to stay within 20% of optimal and has quite acceptable running times even for 100,000 cities. ------------------------------ Date: Fri, 9 Oct 1987 15:16 PDT From: Marty Tenenbaum Subject: Seminar - AI in Manufacturing (SU) FIRST-CUT: A Knowledge-based CAD/CAM System for Concurrent Product and Process Design Prof. Mark Cutkosky (ME) Friday, Oct. 16 at 3:30 pm. Terman 556 Abstract: FIRST-CUT is a novel CAD/CAM system for rapid prototyping of mechanical parts. Parts are designed interactively by graphically composing a high-level plan for producing them. A plan consists of abstract machining operations, such as "make hole" or "make pocket". As each operation is added, expert systems check feasibility and fill in details (e.g., whether a hole should be drilled, milled, or bored.) Also, a solid modeler incrementally simulates the plan to detect geometric interference problems and to enable the designer to visualize the part taking shape. Completed plans are compiled into NC-code and run on a table-top milling machine to physically instantiate the design. A second part of the project is concerned with monitoring the actual execution of process plans on specially instrumented machine tools, and using the results to refine the knowledge base and produce better plans. A live demonstration will follow the talk. Students seeking an exciting real-world domain for AI research (in areas such as planning, spatial reasoning, knowledge-acquisition and learning, intelligent agents, signal understanding and man-machine communication) are especially invited. ------------------------------ Date: Fri 9 Oct 87 15:47:14-PDT From: Anne Richardson Subject: Seminar - Pengi: An Implementation of a Theory of Activity (SU) Daniel Weise is hosting Phil Agre here at Stanford on Tuesday, October 27 who will be giving the following talk in Bldg. 200, Rm. 303 at 4:15: ***For any questions, please contact Daniel@mojave.*** Pengi: An implementation of a theory of activity Phil Agre MIT Artificial Intelligence Laboratory AI has typically sought to understand the organized nature of human activity in terms of the making and execution of plans. There can be no doubt that people use plans. But before and beneath any plan-use is a continual process of moment-to-moment improvisation. An improvising agent might use a plan as one of its resources, just as it might use a map, the materials on a kitchen counter, or a string tied round its finger. David Chapman and I have been studying the organization of the most common sort of activity, the everyday, ordinary, routine, familiar, practiced, unproblematic activity typified by activities like making breakfast, driving to work, and stuffing envelopes. Our theory describes the central role of improvisation and the inherent orderliness, coherence, and laws of change of improvised activity. The organization of everyday routine activity makes strong suggestions about the organization of the human cognitive architecture. In particular, one can get along remarkably well with a peripheral system much as described by Marr and Ullman and a central system made of combinational logic. Chapman has built a system with such an architecture. Called Pengi, it plays a commercial video game called Pengo, in which a player controls a penguin to defend itself against ornery and unpredictable bees. The game requires both moderately complex tactics and constant attention to opportunities and contingencies. I will outline our theory of activity, describe the Pengi program, and indicate the directions of ongoing further research. ------------------------------ Date: Thu 8 Oct 1987 12:40:14 From: dlm@allegra.att.com Subject: Seminar - Persistence, Intention, and Commitment (AT&T) Title: Persistence, Intention, and Commitment Speaker: Philip R. Cohen Affiliation: SRI International Menlo Park, CA Place: AT&T Bell Laboratories Murray Hill 3D-473 Date: October 8, 1987 1:30 PM. (work done jointly with Hector Levesque) Abstract: This talk explores principles governing the rational balance among an agent's beliefs, goals, actions, and intentions. Such principles provide specifications for artificial agents, and approximate a theory of human action (as philosophers use the term). By making explicit the conditions under which an agent can drop his goals, i.e., by specifying how the agent is _committed_ to his goals, the formalism captures a number of important properties of intention. Specifically, the formalism provides analyses for Bratman's three characteristic functional roles played by intentions, and shows how agents can avoid intending all the foreseen side-effects of what they actually intend. Finally, the analysis shows how intentions can be adopted relative to a background of relevant beliefs and other intentions or goals. By relativizing one agent's intentions in terms of beliefs about another agent's intentions (or beliefs), we derive a preliminary account of interpersonal commitments. ------------------------------ Date: 7 Oct 87 12:26:16 GMT From: ihnp4!erc3bb!may@ucbvax.Berkeley.EDU (M.A.Yousry) Subject: Conference - AI and Statistics at JSM At the August 22-25, 1988 Joint Statistical Meetings (American Statistical Association, Biometric Society, Institute of Mathematical Statistics) in New Orleans, I'll be chairing an invited session, titled "Bridging the Gap, Artificial Intelligence and Statistics," on solving problems using combinations of AI and statistical techniques. Both applied and theoretically oriented papers will be considered. Potential areas might include fault diagnosis, process control, reasoning with uncertainty, ... If you are interested in giving a talk, please send a short abstract to: ...!{ihnp4, allegra}!erc780!may or Mona Yousry, (609) 639-2405 AT&T - ERC PO Box 900 Princeton, NJ 08540 ------------------------------ End of AIList Digest ******************** 14-Oct-87 23:28:30-PDT,16053;000000000001 Mail-From: LAWS created at 14-Oct-87 23:19:42 Date: Wed 14 Oct 1987 23:14-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #235 - Business and Marketing, Neuromorphic Terminology To: AIList@SRI.COM AIList Digest Thursday, 15 Oct 1987 Volume 5 : Issue 235 Today's Topics: Queries - Italian AI & NExpert & Scheme as a First Lisp & Engineer/Scientist Expert Systems & Connection Machine Applications to Vision & AI Workstations & Learning Software & Prolog Shopping, Business - Expert Systems Company Financing & AI Marketing, Neuromorphic Systems - Terminology ---------------------------------------------------------------------- Date: 12 Oct 87 07:26:23 GMT From: imagen!atari!portal!cup.portal.com!tony_mak_makonnen@ucbvax.Berk eley.EDU Subject: european connection I would like to E-mail to people in Italy interested in the kind of things we discuss here . i know there is a connection thru Amsterdam and Turin . Any pointers on how to do it and who to contact first are appreaciated . I speak and write italian fluently and will gladly accept messages in that language ------------------------------ Date: Mon, 12 Oct 87 13:16 MET From: SYS_MS@bmc1.uu.se Subject: NExpert Neuron Data's NExpert system shell should soon be available for the Macintosh. Have anyone out there used it for real development. What is the performance compared to the VAX implementation. Pro's and con's of NExpert? Mats ------------------------------ Date: Mon, 12 Oct 87 09:47:42 -0400 From: howell%community-chest.mitre.org@gateway.mitre.org Subject: Scheme as a first lisp? I'm interested in learning lisp "in my spare time", and I'd prefer to do it on my Sun 3/75. For that reason, I'm thinking about bringing GNU C Scheme up on the Sun. Before I do, I have a few questions (of obvious neophyte level!). Thanks in advance for any responses (please respond directly, I'm not on this list). 1) How solid is GNU Scheme? I'm using GNU EMACS and GNU BISON (== YACC), and I've been really happy with both, so I imagine GNU Scheme is fairly bug-free... 2) How different is Scheme from Common Lisp and Franz? 3) Is it a good idea/bad idea/neutral idea for someone intending to learn lisp to start with Scheme? 4) If I go with Scheme, are there other recomended books/articles in addition to "Structure and Interpretation of Computer Programs" by Abelson and Sussman(s) [Sussmen? sorry...] Thanks for any help. Chuck Howell the MITRE Corporation, Mail Stop Z645 7525 Colshire Drive, McLean, VA 22102 (703) 883-6080 ARPA: howell@mitre.arpa ------------------------------ Date: 14 Oct 87 11:23:00 PDT From: "SEF::BROWER" Reply-to: "SEF::BROWER" Subject: Engineer/Scientist Expert System info We are looking into the possibility of creating an expert system to capture the expertise of engineers/scientists and would appreciate any information anyone has on existing systems of this nature or systems being developed of this nature. My address is: Brower@NWC-143B.ARPA until 19 Oct. 87. After Oct. 19 my address changes to: Brower@NWC.ARPA. Thank you in advance. Roger Brower ------------------------------ Date: 14 Oct 87 05:19:37 GMT From: jason@locus.ucla.edu Subject: Looking for connection machine applications to vision I am currently beginning to look into the area of vision research applied to the connection machine, or other connectionist architectures. I would appreciate any good references and input in this area. Jason Rosenberg jason@CS.UCLA.EDU ------------------------------ Date: Wed, 14 Oct 87 19:40 N From: KOLB%HTIKUB5.BITNET@wiscvm.wisc.edu Subject: help! (ai-workstations) hi out there, we are a (for Holland) pretty old and matured AI&NLP research group, but so is our hardware equipment, which we share with the rest of the university. Now we seem to have the chance of getting some stuff such as workstations on our own. Any recommendations (or - even more useful - warnings)? what we're looking for is an integrated environment with good PROLOG-, LISP- and maybe some object-oriented facilities, but also capable of managing old-fashioned languages such as pascal and C. Good graphics facilities would help, too. Please, reply directly to me. I'm gonna summarize the results for the net, if wanted. Thanx, hap kolb Address: EMAIL: kolb@htikub5.bitnet SNAILMAIL: hap kolb Tilburg University - SLE Postbus 90153 NL-5000 LE Tilburg The Netherlands ------------------------------ Date: Tue, 13 Oct 87 10:12:40 GMT From: Richard White Subject: Query - Learning software The Edinburgh Computing and Social Responsibility (CSR) group are looking for software which may beused or adapted for use in an AI teaching module which will (hopefully!) be offered to Scottish school children in 1988, at least on a trial basis. The module, aimed at 16-18 year olds, is concerned with the study of learning in both human and machines. Areas of interest include induction, discovery and analogy based learning. What we are short of are sample programs which can be used to illustrate some of the problems involved in the study and simulation of these processes. By necessity these have to be simple and relatively small, the target machines being Nimbus's (British IBM-compatible PC). Prolog would be the preferable language, but then we can't be too choosy! Software should be public domain as we are running on a **very** small budget. If anyone knows of anything which might be suitable we would be very grateful to hear about it. Could you Email any replies DIRECT to me. Thanks in anticipation, Richard White (on behalf of CSR group) JANET: R.White@uk.ac.edinburgh ARPA: R.White%uk.ac.ed@nss.ucl.ac.uk UUCP: ...!ukc!ed.ac.uk!R.White ------------------------------ Date: 14 Oct 87 13:46:00 EST From: cugini@icst-ecf.arpa Reply-to: Subject: Prolog shopping I'm doing some serious shopping for an industrial strength Prolog to run under VAX/VMS. The only vendors of which I am aware are: 1. Quintus 2. IF/Prolog (Munich Germany) 3. Prolog-1 from Expert Systems International Desirable features include: 1. nice environment/editor for changing and testing 2. external DB - preferably based on SQL 3. ability to call external languages, eg FORTRAN routines. I'd be happy to hear about any new products, assessments, suggestions, etc. John Cugini ------------------------------ Date: 13 Oct 87 16:09:18 GMT From: jbn@glacier.stanford.edu (John B. Nagle) Subject: Re: Expert Systems Company Financing... Right now, the venture capital community has had it with expert systems. Hambrecht of Hambrecht and Quist, one of the more influential venture capitalists, has been quoted as saying "Artificial intelligence is the most effective means yet invented for separating investors from their money". There are no AI startup success stories yet, remember; nothing comparable to SUN or Lotus has happened. A few companies made it to IPO, but the stocks never took off. Of the companies that received a lot of public attention, the score is as follows. Annual Annual Yesterday P/E High Low Intellicorp 11 1/8 4 1/8 5 3/4 29 Teknowledge 21 5/8 8 13 1/4 loss Symbolics 6 1/8 3 3 1/4 loss Lisp Machines (bankrupt) So forget an expert system startup using the venture capital route until somebody makes it. But venture capital is a fad-driven industry. Neural nets are hot this month. John Nagle [John tells me there's a downbeat article on AI in the latest Forbes. (I hope I got that right.) -- KIL] ------------------------------ Date: 14 Oct 87 16:11:57 GMT From: iscuva!randyg@uunet.uu.net (Randy Gordon) Subject: Re: Expert Systems Company Financing... But... That really doesn't reflect on AI's success. There have been quite a number of wildly sucessful AI projects that I know of, but they are usually buried deep in companies that do other things, and noone talks about them, so they won't lose competitive advantage. None of the pure AI companies really had a chance. All they sold were tools to solve problems, and consulting services. But one tool generates many end products, and theres only so much training you can do before your customer knows as much as you do. Companies that sell end products that use AI techniques(such as Syntelligence, or the thousand and one genetic engineering companies) are doing quite well. So are the ones that use AI as part of a tool to increase productivity or spread expertise, like Dec. If any of those pure AI companies had ANYONE with decent marketing(not sales!) experience, they would have started generating applications, (with tools as a sideline). Theres a HUGE vein of expertise out there to be mined. Many industries lack the will, expertise, or political situation to make use of the knowledge that exists and the AI techniques necessary to utilize it. AI techniques can fulfill needs that are difficult to answer with other technologies. In combination with more ordinary programming techniques, you can provide a demonstratably superior product in many areas. But you have to be answering needs! AI companies don't have problems because they are AI, they have problems because noone in them really understands how to succeed as a business, rather than as a glorified consulting firm. Randy Gordon ------------------------------ Date: Wed 14 Oct 87 22:16:47-PDT From: Ken Laws Reply-to: AIList-Request@SRI.COM Subject: AI Marketing Part of what we are seeing in AI is the evolution from horizontal to vertical marketing. Vertical integration (i.e., applications) had to wait for the horizontal suppliers to develop their machines and software -- with the exception of a few early systems such as Dendral and R1/XCON. The horizontal market has saturated, though, partly because it is much easier to develop a general-purpose system than it is to really understand a customer's applications and needs (in addition to developing an AI system capable of handling previously unsolved problems). Unless some new market opens up -- business, military, educational, or consumer -- the horizontal companies have now sold to everyone interested in buying. The companies that will survive are the ones cultivating vertical markets such as warehousing or the printing industry. In some cases these companies are now offering higher priced software with reduced functionality, but with vocabulary and customer support aimed at a specific industry. In other cases the applied systems have not yet become visible simply because it takes a long time to turn a general tool into a useful tool. Expert systems are not dead; the successful ones are just going through another development cycle. The resulting proprietary systems will be hyped in the trade journals rather than the research journals, and will be part of the commercial woodwork from now on. -- Ken ------------------------------ Date: 11 Oct 87 03:32:10 GMT From: wcalvin@well.UUCP (William Calvin) Reply-to: wcalvin@well.UUCP (William Calvin) Subject: Re: Neural Networks - Pointers to good texts? Best book on neural networks is THE CRUSTACEAN STOMATOGASTRIC GANGLION by Selverston and Moulins (Springer 1987). If you mean neural-like networks, try the Rumelhart et al PARALLEL DISTRIBUTED PROCESSING (MIT Press 1986). We brain researchers sure get tired of hearing neural-like networks referred to as "neural networks", an established subject for 25 years since the days of Limulus lateral inhibition. Calling silicon networks "neural" is about like the hype in the early days when every digital computer was called a "brain" by the media. William H. Calvin University of Washington NJ-15, Seattle WA 98195 wcalvin@well.uucp wcalvin@uwalocke.bitnet ------------------------------ Date: 13 Oct 87 03:30:21 GMT From: sabbath.rutgers.edu!leasure@rutgers.edu (David E. Leasure) Subject: Re: Neural Networks - Pointers to good texts? wcalvin@well.UUCP (William Calvin) writes: > We brain researchers sure get tired of hearing neural-like networks >referred to as "neural networks", an established subject for 25 years since >the days of Limulus lateral inhibition. Calling silicon networks "neural" is >about like the hype in the early days when every digital computer was >called a "brain" by the media. Maybe we could all agree on a more faithful/less ingratiating term? Maybe connectionist processing models or neuromorphic (after Touretzky), or fine-grained parallel processing? (even Rumelhart's Parallel Distributed Processing?) David E. Leasure Rutgers/AT&T [Connectionism is a subset of the neuromorphic approach that uses coarse -- or distributed -- coding instead of single nodes to represent concepts. It's like representing all entities by feature vectors instead of by symbol or name. Fine-grained parallel processing includes new architectures such as the Connection Machine that are not related to neural networks (beyond being ideal simulation substrates). I don't know how PDP differs from any other distributed processing, but the latter includes contract nets and inferential databases. I'm willing to use "neuromorphic", although I'm not sure that any one term can adequately describe this diverse field. The name that sticks, though, will be the one that is most effective in prying money from the military and governmental fuding agencies -- and I suspect that "neural networks" will win precisely because of its misleading connotations. -- KIL] ------------------------------ Date: 12 Oct 87 13:35:59 GMT From: ssc-vax!dickey@beaver.cs.washington.edu (Frederick J Dickey) Subject: Re: Neural Networks - Pointers to good texts? In article <4191@well.UUCP>, wcalvin@well.UUCP (William Calvin) writes: > We brain researchers sure get tired of hearing neural-like networks > referred to as "neural networks", an established subject for 25 years since > the days of Limulus lateral inhibition. I think the above says that "biological" neural nets have been studied as a formal discipline for 25 years and that this great ancestry gives biology prior claim to the term "neural nets". Assuming that this is a correct interpretation, let me make the following observation. In 1943, McCulloch and Pitts published a paper entitled "A logical calculus of the ideas immanent in neural nets". Minsky and Papert (Perceptrons) state that this paper presents the "prototypes of the linear threshold functions". This paper stikes me as clearly being in the "neural net-like" tradition. Now 1987-1943 = 44. Also note that 44 > 25. Therefore, it apears that the "neural net-like" guys have prior claim to the term "neural net". :-). ------------------------------ End of AIList Digest ******************** 15-Oct-87 22:36:01-PDT,16020;000000000001 Mail-From: LAWS created at 15-Oct-87 22:31:29 Date: Thu 15 Oct 1987 22:23-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #236 - Semantics of Flawed Minds To: AIList@SRI.COM AIList Digest Friday, 16 Oct 1987 Volume 5 : Issue 236 Today's Topics: Semantics - Is the Human Mind Flawed? ---------------------------------------------------------------------- Date: Mon, 12 Oct 87 10:35 EST From: "Linda G. Means" Subject: ailist discussion of "flawed minds" In AIList V5 #233, ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) writes: >Factually, we know the mind is flawed because we observe that >it does not do what we expect of it. Okay, let's take as given that the human mind is flawed. If that judgment is the result of reasoning by a human mind (i.e. a flawed mind), how can we take the judgment to be true? Seems not unlike the paradox which arises from the statement, "Everything I say is a lie". Linda G. Means GM Research Laboratories means%gmr.com@relay.cs.net [Ah, but that's a far cry from "Some things I say are lies." -- KIL] ------------------------------ Date: 12 Oct 87 17:12:39 GMT From: ucsdhub!jack!man!sdsu!caasi@sdcsvax.ucsd.edu (Richard Caasi) Subject: Re: Is the human mind flawed? If the human mind was flawless we wouldn't be debating this issue. To determine how flawed the human mind is we need to first define the characteristics of a flawless or perfect mind. Any suggestions? It certainly shouldn't have the limitations of Turing machines, that is, it should be able to "solve" non-computable functions non- algorithmically. Given perfect information as input, its output should be likewise perfect, right? Or perhaps its output should always be perfect regardless of how imperfect or incomplete its inputs are. (Whcih violates the CS law of Garbage In Garbage Out) Drawing an analogy with ideal operational amplifiers in electronics, the perfect mind can be characterized by infinite memory, zero learning time, zero search and recall time, sensory perception with infinite bandwidth (flat frequency response from negative infinity to positive infinity), zero computation time, and knowledge of future inputs, etc., etc. (What do we have - God?) Question: Does such a mind exist or is nothing perfect in the real world? ------------------------------ Date: 12 Oct 87 20:12:26 GMT From: pioneer!eugene@ames.arpa (Eugene Miya N.) Subject: Re: Goal of AI: where are we going? In article <578@louie.udel.EDU> montgome@udel.EDU (Kevin Montgomery) writes: >>> In article <2281@umn-cs.UUCP>, ramarao@umn-cs.UUCP (Bindu Rama Rao) writes: >>> > Is the Human mind flawed? >C'mon guys, lighten up for a sec. Flawed implies a defect from it's >design. Therefore, if someone's mind doesn't do what it's designed Having read the postings which followed this, consider that the human eye has many blind spots, the largest where the optic nerve is and many smaller ones. The ear isn't perfect either. Also consider how we can be fooled by Necker illusions, visual, verbal, auditory, etc. Flawed many be too strong a word. Is the greater "mind" be flawed if it's components and inputs are "flawed?" I prefer the "Just is" hypothesis. On emotions: you may have something there, but AI people are not the people to answer that question. A fellow I corresponded with on AI-Digest a while noted he had a difficult time writing a Social Worker expert system. Harder to dish out artificial compassion than artificial Discrimination. From the Rock of Ages Home for Retired Hackers: --eugene miya NASA Ames Research Center eugene@ames-aurora.ARPA "You trust the `reply' command with all those different mailers out there?" "Send mail, avoid follow-ups. If enough, I'll summarize." {hplabs,hao,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene ------------------------------ Date: 12 Oct 87 20:23:46 GMT From: gatech!pyr!kludge@rutgers.edu (Scott Dorsey) Subject: Re: Is the human mind flawed? If a thing is not perfect, it is flawed def. flaw The human mind is a thing if it weren't, we wouldn't talk about it Nothing is perfect My mother said this ------------------------------------------------------------------------------- The human mind is flawed QED. -- Scott Dorsey Kaptain_Kludge SnailMail: ICS Programming Lab, Georgia Tech, Box 36681, Atlanta, Georgia 30332 Internet: kludge@pyr.gatech.edu uucp: ...!{decvax,hplabs,ihnp4,linus,rutgers,seismo}!gatech!gitpyr!kludge ------------------------------ Date: 12 Oct 87 15:11:38 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Is the human mind flawed? In article <17489@yale-celray.yale.UUCP>, krulwich@gator..arpa (Bruce Krulwich) writes: > In article <1368@houdi.UUCP> marty1@houdi.UUCP (M.BRILLIANT) writes: > >Factually, we know the mind is flawed because we observe that it does > >not do what we expect of it. > > If I expect my car to take me to the moon and it doesn't, is it > flawed?? No, rather my expectation of it is wrong. Similarly, we > shouldn't say that the mind is flawed until we're sure that our > definition of "intelligence" is perfect. There's a subtlety here. Your car is obviously not designed to go to the moon; it won't come near trying. But I suggested that your car should take you "from Pittsburgh to Atlanta" without bursting into flame. That's not an unreasonable expectation, because, though it probably wasn't designed for those particular roads, cars like it usually do it successfully. Similarly, if I usually go through interviews without "bursting into flame," I expect to be able to do it regularly, and if once I screw up, I have to conclude that there is a flaw somewhere. > > As a hypothesis, we can test the idea > >that it is flawed because of the action of what we call emotions. > > Why do you assume that emotions are a flaw?? Just maybe emotions are > at the core of intellegence, and logic is just a side issue. Note, please. I did not "assume that emotions are a flaw." First, I argued that there was a flaw, and though that argument was challenged, my reliance on that argument is obviously "why" I went on to the next step. Second, I obviously did not "assume" anything about emotions; I offered a hypothesis about emotions. "Why" I offered that hypothesis is that it was suggested by an article I quoted: == > Is the mind flawed just because humans make decisions based on == > their emotional involvement? .... > If you think that emotions motivate all human activity, why do you > dismiss emotions as a flaw in the mind?? It seems to me that human > activity is a lot more "intelligent" than any AI system as of yet. Clearly I did not dismiss anything. Quoting again from my article: == > Let's not hastily dismiss the human mind as flawed. == == Who's dismissing it? I know my car is flawed, but I can't afford to == dismiss it. I'm not dismissing my mind either. How could I? :-) Without trying to embarrass anybody, I would like to ask whether Mr. Krulwich thought he was answering logically, and, if so, whether his expectation that he could do so was any more reasonable than the hypothetical expectation that his car could take him to the moon. I think we try to do things with our minds that they can not successfully do. Even if the flaw is in the expectation, the expectation is created by the mind, so to argue that the flaw is not in the mind requires great subtlety. (I am sure many readers will find my argument flawed.) I might suggest that Mr. Krulwich answered more emotionally than logically, but that statement would not only introduce "emotion" as an undefined term, but also invite us to "dismiss" what seem to be some vital mental processes. Just as physicians accept the human body for what it is, without embarrassment, so should we accept the human mind. Physically, all human bodies are different, and none are perfect. Why then should anyone insist that the mind is unflawed? M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 13 Oct 87 12:47:26 GMT From: PT.CS.CMU.EDU!SPICE.CS.CMU.EDU!spe@cs.rochester.edu (Sean Engelson) Subject: What the hell does flawed mean, anyway? Could someone please define flawed, as it applies (or may apply) to the mind? Flawed with respect to the performance of what action? Formal logic? Aristotelian logic? Type theory? NP-complete computations? Getting emotional? You need referents! I think that most people are just talking past each other, as they are using different referents. I am not getting involved yet, as I don't think that I know what referents are appropriate---if anyone thinks they know: What are they??? -Sean- ------------------------------ Date: 10 Oct 87 10:27:50 GMT From: ihnp4!homxb!mtuxo!mtune!codas!killer!usl!khl@ucbvax.Berkeley.EDU (Calvin K. H. Leung) Subject: Re: Goal of AI: where are we going? (the right way?) In article <1270@isl1.ri.cmu.edu> cycy@isl1.ri.cmu.edu (Christopher Young) writes: > I do believe that there is some mechanism to minds (or perhaps a variety of > them). One reason why I am interested in AI (perhaps this is very Cog. Sci. > of me, actually) is because I think perhaps it will help elucidate the ways > in which the human mind works, and thus increase our understanding of human > behaviour. I agree with the idea that there must be some mechanisms that our minds are using. But the different reasoning methods (proba- bilistic reasoning, for instance) that we are studying in the area of AI are not the way one reasons: we never use the Bayes' Theorem in our thinking process. The use of those reasoning methods, in my point of view, will never help increase our under- standing of human behavior. Because our minds just don't work that way. Calvin K H Leung -- Calvin K. H. Leung USL P.O. Box 41821 Lafayette, LA 70504 khl@usl.usl.edu.csnet 318-237-7128 ------------------------------ Date: 14 Oct 87 15:47:09 GMT From: ihnp4!homxb!genesis!odyssey!gls@ucbvax.Berkeley.EDU (g.l.sicherman) Subject: Re: Flawed human minds > Let's draw an analogy. You are driving an X-Brand car from Pittsburgh to > Atlanta and halfway there it bursts into flame. Without knowing how the > car works you can conclude it was flawed. > > Mr X. goes to an employment interview and gets angry or flustered and > says something that causes him to be rejected. Without knowing how his > mind works you can conclude it was flawed. And you could be wrong. Most likely Mr. X. didn't want the job after all. He only wanted you to think he wanted the job. Give him credit for some intelligence! Of course Mr. X. is flawed from the company's point of view. But he's flawed from his own point of view only if he can get what he wants and doesn't. When this happens, the problem is not emotions but habits. > Factually, we know the mind is flawed because we observe that it does > not do what we expect of it. By this criterion, we are all flawed. It brings to mind the one and only law in J. B. Cabell's land of Philistia: "Do what seems to be expected of you." -- Col. G. L. Sicherman ...!ihnp4!odyssey!gls ------------------------------ Date: Wed 14 Oct 87 21:36:57-PDT From: Ken Laws Reply-to: AIList-Request@SRI.COM Subject: Re: Flawed human minds I haven't read Cabell, but I find the quote interesting. I've been saying something similar to family and friends for several years now -- people (esp. children) do what is expected of them, not what is demanded of them. If teachers understood this they could get far more out of their students. Expectation sets up a feedback loop in which the teacher does whatever is necessary to elicit the desired behavior, whereas requests, demands, etc., are events rather than processes. Similar feedback loops are operative in the "lead" of a good dancer or the "ki" of a martial artist. -- Ken ------------------------------ Date: 14 Oct 87 00:19:54 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: What the hell does flawed mean, anyway? In article <160@PT.CS.CMU.EDU>, spe@SPICE.CS.CMU.EDU (Sean Engelson) writes: > > Could someone please define flawed, as it applies (or may apply) to > the mind? Flawed with respect to the performance of what action? > Formal logic? Aristotelian logic? Type theory? NP-complete > computations? Getting emotional? .... All of the above. > ... You need referents! I think that > most people are just talking past each other, as they are using > different referents. I am not getting involved yet, as I don't think > that I know what referents are appropriate---if anyone thinks they > know: What are they??? I claim that with respect to any referent the mind is flawed. If any reader can define any referent with respect to which the mind is perfect, I will admit my argument is flawed. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 15 Oct 87 00:35:32 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: Flawed human minds In article <331@odyssey.ATT.COM>, gls@odyssey.ATT.COM (g.l.sicherman) writes (quoting from something I wrote): > > Let's draw an analogy. You are driving an X-Brand car from Pittsburgh to > > Atlanta and halfway there it bursts into flame. Without knowing how the > > car works you can conclude it was flawed. > > > > Mr X. goes to an employment interview and gets angry or flustered and > > says something that causes him to be rejected. Without knowing how his > > mind works you can conclude it was flawed. > > And you could be wrong. Most likely Mr. X. didn't want the job after > all. He only wanted you to think he wanted the job. Give him credit > for some intelligence! > > Of course Mr. X. is flawed from the company's point of view. But he's > flawed from his own point of view only if he can get what he wants and > doesn't. When this happens, the problem is not emotions but habits. Also flawed from Mr. X's point of view. Sicherman argues that X only seemed to get angry or flustered, in order to make sure the company didn't make him an offer, because during the interview he decided he didn't want a job with them. If I attributed Mr. X's actions to intelligence I would expect him to conclude gracefully, let them make an offer, and reject the offer, without making a bad impression on somebody who later might be in a position to offer him a job in another company. And I don't care whether you blame emotions or habits. > > Factually, we know the mind is flawed because we observe that it does > > not do what we expect of it. > > By this criterion, we are all flawed.... That's exactly what I meant. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ End of AIList Digest ******************** 15-Oct-87 22:59:15-PDT,18465;000000000000 Mail-From: LAWS created at 15-Oct-87 22:56:58 Date: Thu 15 Oct 1987 22:54-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #237 - Seminars, Connectionist Course, Conference To: AIList@SRI.COM AIList Digest Friday, 16 Oct 1987 Volume 5 : Issue 237 Today's Topics: Seminars - The Logical Foundations of Evidential Reasoning (SRI) & The Matrix of Biological Knowledge (BBN) & PROLOG and AI Applications - A European Perspective (UNISYS) & Non-Deterministic Lisp (SRI) & OB1: A Prolog-Based Object-Oriented Database (UNYSIS), Course - Connectionist Summer School, Conference - Computers in Engineering ---------------------------------------------------------------------- Date: Wed, 14 Oct 87 16:10:01 PDT From: seminars@csl.sri.com (contact lunt@csl.sri.com) Subject: Seminar - The Logical Foundations of Evidential Reasoning (SRI) SRI COMPUTER SCIENCE LAB SEMINAR ANNOUNCEMENT: THE LOGICAL FOUNDATIONS OF EVIDENTIAL REASONING Enrique H. Ruspini Artificial Intelligence Center SRI International Monday, October 19 at 4:00 pm SRI International, Computer Science Laboratory, Room IS109 The approach proposed by Carnap for the development of logical bases for probability theory is applied to formal structures that are based on epistemic logics. Epistemic logics are modal logics introduced to deal with issues that are relevant to the state of knowledge that rational agents have about the real world. The use of epistemic logics in problems of analysis of evidence is justified by the need to distinguish among such notions as the state of a real system, the state of knowledge possessed by rational agents, and the impact of information on that knowledge. Carnap's method for generating a universe of possible worlds is followed using an enhanced notion of possible world that encompasses descriptions of knowledge states. Within such generalized or epistemic universes, several classes of sets are identified in terms of the truth-values of propositions that describe either the state of the world or the state of knowledge that rational agents have about it. Probabilities defined over certain subsets of the epistemic universe are then shown to have the properties of the belief and basic probability assignment functions of the Dempster-Shafer calculus of evidence. Furthermore, extensions of a probability function defined over epistemic subsets (representing different states of knowledge) to truth-sets (representing true states of the real world) must satisfy the interval probability bounds derived from the Dempster-Shafer theory. These bounds correspond to the classical notions of lower and upper probability and are the best possible, given a specific state of knowledge. Finally, the problem of combining the knowledge state of several rational agents is also treated by consideration of epistemic structures. The result of this analysis is a general formula for the integration of evidence. From this formula and certain probabilistic independence assumptions, the rule of combination of Dempster is easily derived. The meaning of these independence assumptions is made explicit through the insight provided by the formal structures that are used to represent knowledge and truth. NOTE FOR VISITORS TO SRI: Please arrive at least 10 minutes early in order to sign in and be shown to the conference room. SRI is located at 333 Ravenswood Avenue in Menlo Park. Visitors may park in the visitors lot in front of Building A (red brick building at 333 Ravenswood Ave) or in the conference parking area at the corner of Ravenswood and Middlefield. The seminar room is in the International Building -- the white concrete structure on Ravenswood to the East (left) of Building A. Visitors should sign in at the International Building reception --- up the steps into the courtyard and on the left. IMPORTANT: Visitors from Communist Bloc countries should make the necessary arrangements with Fran Leonard (415-859-4124) in SRI Security as soon as possible. ------------------------------ Date: Tue 13 Oct 87 15:27:56-EDT From: Marc Vilain Subject: Seminar - The Matrix of Biological Knowledge (BBN) BBN Science Development Program Joint Biotech and AI Seminar Series Lecture "The Matrix of Biological Knowledge" Kimberle Koile BBN Labs (KKOILE@G.BBN.COM) BBN Labs 10 Moulton Street 2nd floor large conference room 10:30 am, Thursday October 15th The body of experimental data in the biological sciences is immense and growing rapidly. Its volume is so extensive that computer methods, possibly straining the limits of current technology will be necessary to organize the data. Moreover, it seems highly likely that there are a significant number of as yet undiscovered ordering relations, new laws, and predictive models embedded in the mass of existing information. To employ this body of information productively, it will be useful to create an extensive data/knowledge base, "the matrix of biological knowledge," structured to provide a conceptual framework by the laws, models, empirical generalizations, and physical foundations of the modern biological sciences. --- from a Santa Fe Institute press release This talk will describe preliminary efforts to define and prototype parts of the Matrix. These efforts took place at a summer workshop that was organized as a result of a National Academy of Sciences report published in 1985, "Models for Biomedical Research: A New Perspective." The workshop, sponsored by the Santa Fe Institute with support from NIH, DOE, and several commercial companies, was attended by fifty scientists from a variety of biology and computer subdisciplines. Note: A related talk on the Matrix will be given Friday morning (announcement forthcoming) by Prof. Harold Morowitz of the Department of Biophysics and Biochemistry at Yale University. Prof. Morowitz chaired the Committee on Models for Biomedical Research, which produced the above mentioned report, and co-chaired the Workshop on the Matrix of Biological Knowledge. ------------------------------ Date: Wed, 14 Oct 87 15:50:48 EDT From: finin@bigburd.PRC.Unisys.COM (Tim Finin) Subject: Seminar - PROLOG and AI Applications - A European Perspective (UNISYS) AI Seminar UNISYS Knowledge Systems Paoli Research Center Paoli PA PROLOG AND AI APPLICATIONS - A EUROPEAN PERSPECTIVE Raf Venken BIM Prolog Raf Venken, manager for BIM Prolog Research and Development, will be visiting Logic Based Systems on Monday, October 19th. BIM is a high performance Prolog which runs on UNIX-based SUN workstations as well as VAXES under VMS, UNIX 4.2, and ULTRIX. BIM is involved in joint research efforts with various universities throughout Europe and is a member of ESPRIT (the "European MCC"). BIM has also contributed to LOQUI, a large natural language project. BIM claims to be the fastest general purpose Prolog system currently available on the market. BIM includes "the first successful attempt to include more intelligent debugging aids into the [Prolog] system" and a "PARTIAL EVALUATION system which optimizes Prolog programs by source-to-source transformations." BIM has also "extended the Prolog language with the concept of MODULES to allow the easy development of very large systems." The talk will cover the philosophy and strategy behind BIM Prolog, discuss current ESPRIT projects including a large NLP system, and speculate about the future. 11:00am, Monday, October 19th Cafeteria Conference Room - if you are interested in attending, please send - - mail to finin@prc.unisys.com or call 215-648-7446 - ------------------------------ Date: Wed, 14 Oct 87 11:54:41 PDT From: Amy Lansky Subject: Seminar - Non-Deterministic Lisp (SRI) DEPENDENCY-DIRECTED BACKTRACKING IN NON-DETERMINISTIC LISP Ramin Zabih (RDZ@SUSHI.STANFORD.EDU) Computer Science Department Stanford University 11:00 AM, MONDAY, October 19 SRI International, Building E, Room EJ228 Dependency-directed backtracking is a strategy for solving generate-and-test search problems. Pure Lisp extended with McCarthy's non-deterministic operator AMB is an elegant language for expressing such problems. I will describe how to automatically provide dependency-directed backtracking in SCHEMER, a non-deterministic Lisp dialect. It is also possible for SCHEMER to automatically provide other search strategies than dependency-directed backtracking. In fact, SCHEMER can support a large class of solution methods. I will show that SCHEMER programs can make use of any algorithm for determining the satisfiability of a propositional formula in Conjunctive Normal Form. This is joint work with David McAllester. VISITORS: Please arrive 5 minutes early so that you can be escorted up from the E-building receptionist's desk. Thanks! ------------------------------ Date: Thu, 15 Oct 87 15:12:54 EDT From: finin@bigburd.PRC.Unisys.COM (Tim Finin) Subject: Seminar - OB1: A Prolog-Based Object-Oriented Database (UNYSIS) AI Seminar UNISYS Knowledge Systems Paoli Research Center Paoli PA OB1: A PROLOG-BASED OBJECT-ORIENTED DATABASE Benjamin Cohen SRI International David Sarnoff Research Center Princeton NJ 8540 In this talk I describe OB1, an object-oriented database facility. OB1 is a hybrid query language that incorporates most of the features of relational query languages plus "view/objects" that allow sets as values and recursive views. OB1 is implemented in Quintus Prolog and includes server facilities that allow C & Fortran clients to query an OB1 server over a SUN network. OB1 also has a graphics Entity/Relationship data modeling editor used to design OB1 databases. Ben will be here friday, October 23, from lunch time till 5 - I suppose the talk would start around 1:30 or 2:00 2:00pm Friday, October 23 Cafeteria Conference Room - if you are interested in attending, please send - - mail to finin@prc.unisys.com or call 215-648-7446 - ------------------------------ Date: Wed 14 Oct 87 03:18:33-EDT From: Dave.Touretzky@C.CS.CMU.EDU Subject: Course - Connectionist Summer School THE 1988 CONNECTIONIST MODELS SUMMER SCHOOL ORGANIZER: David Touretzky ADVISORY COMMITTEE: Geoffrey Hinton, Terrence Sejnowski SPONSORS: The Sloan Foundation; AAAI; others to be announced. DATES: June 17-26, 1988 PLACE: Carnegie Mellon University, Pittsburgh, Pennsylvania PROGRAM: The summer school program is designed to introduce young neural network researchers to the latest developments in the field. There will be sessions on learning, theoretical analysis, connectionist symbol processing, speech recognition, language understanding, brain structure, and neuromorphic computer architectures. Students will have the opportunity to informally present their own research and to interact closely with some of the leaders of the field. PARTIAL LIST OF FACULTY: Yaser Abu-Mostafa (Caltech) James McClelland (Carnegie Mellon) Dana Ballard (Rochester) David Rumelhart (Stanford) Andrew Barto (U. Mass.) Terrence Sejnowski (Johns Hopkins) Gail Carpenter (Boston U.) Paul Smolensky (UC Boulder) Scott Fahlman (Carnegie Mellon) David Tank (AT&T Bell Labs) Geoffrey Hinton (Toronto) David Touretzky (Carnegie Mellon) George Lakoff (Berkeley) Alex Waibel (ATR International) Yann Le Cun (Toronto) others to be announced EXPENSES: Students are responsible for their meals and travel expenses, although some travel assistance may be available. Free dormitory space will be provided. There is no tuition charge. WHO SHOULD APPLY: The summer school's goal is to assist young researchers who have chosen to work in the area of neural computation. Participation is limited to graduate students (masters or doctoral level) who are actively involved in some aspect of neural network research. Persons who have already completed the Ph.D. are not eligible. Applicants who are not full time students will still be considered, provided that they are enrolled in a doctoral degree program. A total of 50 students will be accepted. HOW TO APPLY: By March 1, 1988, send your curriculum vitae and a copy of one relevant paper, technical report, or research proposal to: Dr. David Touretzky, Computer Science Department, Carnegie Mellon University, Pittsburgh, PA, 15213. Applicants will be notified of acceptance by April 15, 1988. ------------------------------ Date: Mon, 12 Oct 87 09:10:58 EDT From: decvax!cvbnet!cheetah!rverrill@decwrl.dec.com (Ralph Verrilli) Subject: Conference - Computers in Engineering CALL FOR PAPERS 1988 ASME INTERNATIONAL COMPUTERS IN ENGINEERING CONFERENCE AND EXHIBITION SAN FRANCISCO HILTON SAN FRANCISCO, CALIFORNIA July 31 - August 3, 1988 REAL WORLD APPLICATIONS OF EXPERT SYSTEMS AND ARTIFICIAL INTELLIGENCE The theme for the 1988 ASME International Computers in Engineering Conference will focus on the emerging applications of expert systems and artificial intelligence. This conference and exhibition provides a forum for engineers, managers, researchers, vendors, and users to discuss relevant issues, and to present ideas on computer technology and its impact on the engineering workplace. Over 80 papers and panel sessions are planned covering a broad spectrum of technical computing and computers in the engineering community. The topics covered will encompass: computer aided design and manufacturing, computer simulation, robotics, interactive graphics, finite element techniques, microprocessors, computers in educations, expert systems, and artificial intelligence. Papers are solicited in all areas related to the application, development, research, and education with computers in mechanical engineering. Contributions in the form of full-length papers or extended abstracts are solicited. Accepted papers will be published in the bound Conference Proceedings. Full length papers of special note will be reviewed after the conference for publication in the Society's magazine "Computers in Mechanical Engineering (CIME)". The annual event is sponsored by the Computers in Engineering Division of the American Society of Mechanical Engineers (ASME). San Francisco is the site of this years conference. DEADLINES : Submission of three copies of draft contributions (paper or extended abstract) November 30, 1987 Notification of acceptance to authors February 15, 1988 Submission of author-prepared mats April 1, 1988 For the following technical areas please send papers to the respective program chairmen : { Computer Aided Manufacturing, Computer Simulation, Turnkey CAD/CAM, Integration of CAD and CAM, Computer Aided Testing, Computer Aided Design, Interactive Graphics : Dr. Donald Riley Dept. of Mechanical Engineering University of Minnesota 111 Church Street Minneapolis, MN 55455 612-625-0591/1809 } { Artificial Intelligence, Knowledge Based Systems : Mr. M.F. Kinoglu AI and Expert Systems Group Control Data Corporation 1450 Energy Park Drive Saint Paul, MN 55108 612-642-3817 } { Microprocessors, Robotics, Special Purpose Computers, Man-Machine Interfaces : Mr. David W. Bennett Battelle Pacific Northwest Labs P.O. Box 999 Richland, WA 99352 509-375-2159 } { Robotics in Education, Teaching CAD in Higher Education, University - Industry Collaboration, Microcomputers in the Classroom, Computer-Aided Learning : Dr. Gary Kinzel Ohio State University Dept. of Mechanical Engineering 206 West 18th Street Columbus, Ohio 43210 614-292-6884 } { Finite Element Techniques, Software Standards, Computational Geometry : Dr. Kumar K. Tamma Dept of Mechanical Engineering and Aerospace Engineering West Virginia University Morgantown, West Virginia 304-293-4111 } { Computers in Energy Systems, Computational Fluid Dynamics, Computational Heat Transfer, Combustion Modelling, Process Control : Dr. Ahmed A. Busaina Dept. of Mechanical Engineering Clarkson University Potsdam, New York 315-268-6574 } Topics not in the above categories contact Technical Program Chairman : Mr. Edward M. Patton US Army Ballistic Research Lab Aberdeen Proving Grounds, MD 21005 301-278-6805 ------------------------------ End of AIList Digest ******************** 18-Oct-87 23:32:38-PDT,20592;000000000000 Mail-From: LAWS created at 18-Oct-87 23:21:23 Date: Sun 18 Oct 1987 23:14-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #238 - Fault Diagnosis, Financing, AI Successes To: AIList@SRI.COM AIList Digest Monday, 19 Oct 1987 Volume 5 : Issue 238 Today's Topics: Queries - Fire, Women, and Dangerous Things & RB5X Robot & Introductory Books on Lisp & "Eliza-Like" People Stories & Net Mail to UK, Application - Expert Systems for Fault Diagnosis, Humor - "Eliza-Like" People Stories, Education - Learning-Software Reference, Business - Expert Systems Company Financing, Opinion - The Success of AI ---------------------------------------------------------------------- Date: 13 Oct 87 08:27:52 GMT From: cunyvm!unknown%psuvm.bitnet@ucbvax.Berkeley.EDU Subject: Fire, Women, and Dangerous Things I finally located a copy of George Lakoff's "Women, Fire, and Dangerous Things: What Categories Reveal about the Mind" and a quick look through the chapters seems to indicate that it is an interesting synthesis. Anyone out there read it already? I am interested in what those with more experience in AI than I have think of Lakoff's approach to categorization. John M. Ford fordjm@byuvax "The thing I hate about psychologists is that they are always *classifying* everyone..." ------------------------------ Date: Fri 16 Oct 87 15:01:26-PDT From: Matt Heffron Subject: Query: "Not quite a toy" Robot For lack of any more obvious place to ask this... I have been given "custody" of a very sick "Not quite a toy" robot. (The R2D2 sort-of clone that is often seen attracting attention to vendors at trade shows.) The robot was given to the principal of a small private school who gave it to me in the hope that I can repair it, so they can use it at their fund raising events. The problem is that although it has a manufacturer's name, city and model/serial number, the manufacturer (RB Robot Corporation of Golden, Colorado model RB5x) doesn't exist (according to the phone company). Does anyone know someone who might have another of RB's robots (and have schematics, or any documentation at all)? I'd rather fix what's there if I can, instead of rebuilding the complete electronics sub-system from scratch. Thanks in advance, Matt Heffron BEC.HEFFRON@ECLA.USC.EDU PS. I know that this is sort of a shot in the dark, but "netlanders" are sufficiently knowledgeable that if anyone would know how to get some info for this, they would. Thanks. MH ------------------------------ Date: Fri, 16 Oct 87 11:03:46 PDT From: glasgow@marlin.nosc.mil (Michael G. Glasgow) Subject: Introductory books on Lisp I am new to AIList and AI programming and want to learn Lisp. I have been looking through Steele's book, Common Lisp", and have discovered that this is more of a reference manual than a beginners guide. What I am wondering is if anyone can give me the names of some good introductory Lisp books to get me started. Thanks in Advance, michael Net: glasgow@marlin.nosc.mil Reallife: NOSC - Code 423 271 Catalina Blvd. San Diego, CA 92152-5000 ------------------------------ Date: 16-OCT-1987 17:09:17 From: HANCOXPJ@MAIL.ASTON.AC.UK Date: 16-Oct-1987 16:56 BST Subject: Request for information From: Dr P J Hancox Dept: Computer Science Tel No: 021 359 3611 X4652 TO: Remote Addressee ( _POST IKBSBB@RL.VD ) TO: Remote Addressee ( _POST AILIST-REQUEST%COM.SRI.STRIPE@ CC: Remote Addressee ( _KIRK::WOONIMA ) I'm constructing a qualitative model for financial analysis and planning for my PhD which I should finish in late 1988. I intend to supplement this model with quantitative data held in a Financial Modelling system(FPS/EPS2), to handle ambuiguous situations. I am therefore interested in any work on: qualitative models for financial analysis automated interfacing between expert systems and financial modelling systems. Is there anyone out there doing or interested in similar work? Irene Woon JANET: woonimy@uk.ac.aston.kirk uucp: ...seismo!mcvax!ukc!astonk!woonimy phone: + 44 21 359 3611 extn 4272 Snailmail: Department of Computer Science and Applied Mathematics Aston University Birmingham. B4 7ET United Kingdom ------------------------------ Date: 15 Oct 87 18:18:12 GMT From: topaz.rutgers.edu!josh@rutgers.edu (J Storrs Hall) Subject: "Eliza-like" people stories I'm doing a paper on the relation of human conscious processes to those of AI programs. I'm looking for stories which illustrate the extent to which apparent human intelligence may actually consist of Eliza-like verisimilitude. Example: Customer: I'd like to return this pair of shoes. They're both left shoes and one is two sizes smaller than the other. Clerk: We don't take returns. How do we know you haven't worn them? [from Reader's Digest] Please send stories to me rather than the net... --JoSH ------------------------------ Date: Fri, 16 Oct 87 14:04:55 MDT From: yorick%nmsu.csnet@RELAY.CS.NET Subject: Net mail to UK Can any informed person out there tell me what is going on with e-mail to the UK? There seems to have been some radical change in the last month or so completely independent of the general change in destination formats in the US (e.g. com, edu, gov, cs.net and all that). The standard final component @ucl-cs.arpa no longer seems to work as it has for a decade or so. A new format is occuring in UK originating messages, in this list and elsewhere, namely @nss.cs.ucl.ac.uk but that doesnt seem to work as a destination from the US, moreover it is highly confusing as it seems to import the internal UK JANET symbols (ac.uk) into the arpanet address. Since it doesnt work maybe it doesnt matter. There doesnt seem must use asking UK people as they dont know why they can get out as usual but people aree having more trouble reaching them. Another thing is that the preceding part of the UK addresses (e.g. essex.ac.uk) in bloggs%essex.ac.uk@nss.whatever is now being quoted randomly in orogianting headers in both orders e.g. essex.ac.uk and uk.ac.essex. It always used to be the former. Maybe someone in the UK knows what is going on there as it seems that it must be UK rather that US stupidity. I'd be really grateful for any wizard who can tell me either what's going on, or, better still, how to get back to standard reliable transatlantic e-mail. Yorick Wilks. [NSS.CS.UCL.AK.UC seems to have dropped out of the host table at the moment. There is an entry for NS2..., but the socket number differs from [128.41.9.3] and so must be something other than a typo. UCL also has entries for VTEST, TUNNEL, SAM, and TIGER, but not for UCL-CS. As for the problems of the last month, I am beginning to get some leads. The new Arpanet system insists that addresses contain only official host names, and Arpanet hosts will convert aliases to socket numbers if they can't determine the official names. Many Unix systems, though, are still willing to send and receive host aliases, but will reject mail to socket numbers (since such mail in the past has been associated with mailer loops). Mail from an Arpanet host to a Unix host may therefore fail if the Arpanet host tables are not set up exactly right. Many Unix postmasters are not aware of this glitch, or perhaps do not know how to verify and correct the Arpanet host tables. I presume that this has been the case with UCL, although I don't know the nature of their system. I will attempt to get things straightened out if I can get a message through to UCL. -- KIL] ------------------------------ Date: 16 Oct 87 15:09:58 GMT From: moss!erc3bb!may@RUTGERS.EDU (M.A.Yousry) Subject: Re: Engineer/Scientist Expert System info In article <8710150650.AA02610@ucbvax.Berkeley.EDU>, brower%sef.DECnet@NWC-143B.ARPA.UUCP writes: > > We are looking into the possibility of creating an expert system to > capture the expertise of engineers/scientists and would appreciate any > information anyone has on existing systems of this nature or systems being > developed of this nature. > We are working on an expert system to find root causes of fault in a manufacturing process. We are using a statistical system to filter the observations coming from the process, such as defects, or analog measurments..then a rule based system is triggered by the statistical output, uses the engineering knowledge and expertise to find root causes of faults in the process. Bob Parry ihnp4!erc780!bep Mona Yousry ihnp4!erc780!may ------------------------------ Date: 16 Oct 87 14:00:48 GMT From: ihnp4!homxb!vertigo!roller@ucbvax.Berkeley.EDU (P.MICHAELIS) Subject: Re: "Eliza-like" people stories > I'm doing a paper on the relation of human conscious processes to > those of AI programs. I'm looking for stories which illustrate the > extent to which apparent human intelligence may actually consist of > Eliza-like verisimilitude. I know this looks like a portion of a "M*A*S*H" script, but it really did happen this way: YOUNG, OVERWORKED DOCTOR: Why have you come to the hospital? RECENTLY WOUNDED SOLDIER: Shrapnel wounds, sir. YOUNG, OVERWORKED DOCTOR: How long have you been noticing these symptoms? -- Paul Michaelis {AT&T Spine}!vertigo!roller ------------------------------ Date: Thu, 15 Oct 87 13:59:30 EDT From: rapaport@cs.Buffalo.EDU (William J. Rapaport) Subject: induction >From: rw@aiva.edinburgh.ac.UK (Richard White) Subject: Query - Learning software The Edinburgh Computing and Social Responsibility (CSR) group are looking for software which may be used or adapted for use in an AI teaching module ... Take a look at: Robert L. Causey, "Simulations and Experiments in Philosophy of Science," [IBM] Perspectives in Computing, Vol. 7, No. 1 (Spring 1987), pp. 23-33. William J. Rapaport Assistant Professor Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3181 uucp: ..!{ames,boulder,decvax,rutgers}!sunybcs!rapaport internet: rapaport@cs.buffalo.edu [if that fails, try: rapaport%cs.buffalo.edu@relay.cs.net or: rapaport@buffalo.csnet ] bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: 16 Oct 87 18:08:44 GMT From: faline!sabre!gamma!pyuxp!pyuxv!sr@bellcore.bellcore.com (S Radtke) Subject: Re: Expert Systems Company Financing... In article <810@iscuva.ISCS.COM> randyg@iscuva.UUCP (Randy Gordon) writes: > >That really doesn't reflect on AI's success. There have been quite a number >of wildly sucessful AI projects that I know of, but they are usually buried >deep in companies that do other things, and noone talks about them, so >they won't lose competitive advantage. Come on, Randy, let's hear what the wildly successful AI projects were. Most success stories I've heard had to be discounted considerably. They tend to be stories about developments that are full of promise, rather than systems that pay dividends or work for a living. The reports from DEC about Xcon, for instance, did not include bottom line calculations that include system development cost retrieval and maintenance cost, though such support systems are part of the infrastructure and are hard to show as profit centers. Steve Radtke pyuxv!sr ------------------------------ Date: 17 Oct 87 18:26:55 GMT From: imagen!atari!portal!cup.portal.com!barry_night-person_stevens@uc bvax.Berkeley.EDU Subject: you CAN get funding for expert systems activities It's true that the companies started around the large, LISP-based AI machines have not done well. I have recently finished a survey of 179 companies buying and using expert s system tools. Also studied - several vendor companies for expert system products. In short, the big machines aren't what they want -- that's why the companies didn't do well. Computer science-y things arent what they want, either. They are using systems most that are: simple to use, and in English (not in PROLOG) easy to use to access databases, both in PCs and in mainframes easy to interconnect, and to integrate with their corporate data, pgms The smaller, simpler systems are doing well. Also, venture capital firms are cautious about startups. Most prefer to let someone else take the big risks. A few firms, such as Crosspoint and The Sprout Group, will deal with seed. To get funding from a professional source, you need more than a top drawer product idea. You need a quality management team, or the knowledge that one needs to be built; you need a good marketing study, PROVING that a demand for your product exists, and sizing that demand. These, at a minimum. If you also have a good handle on how your venture will work operationally, you are just that much better off. Most of all, you'll need a good estimate of what needs to be done to get your idea into production, and how much it will cost. You also need a top-quality professional technical team to do it with. It may help to realize that you're fighting significant odds. In researching a book I just completed (How to Write A Successful Business Plan, AMACOM) we surveyed 900 venture capital firm and compiled some statistics. only 1 in 2,500 plans that arrive "over the transom" at a VC firm are ever funded. if plans arrive through a trusted associate, 1 in 50 of those plans are funded. Getting funded then becomes a process: put together a top-quality, unique product idea; get a quality, experienced management and professional team together; PROVE THAT YOUR PRODUCT WILL SELL, preferably by actual sales; put a set of thorough financial projections for revenue and ... you get the picture by now. Most people who put together a business plan and try and get funding will probably not get funding. Those companies that have followed the steps I've hinted at DO have a shot at funding. There is a fund being set up just for the funding of companies in the AI area, and that would be a logical place to start. Yes, some of the big-machine companies have failed. Yes, investors have been burned, and most of them are staying away. Yes, there has been too much "smoke and mirrors" about AI. But... investments are STILL being made in expert systems companies. But to get YOUR shot, you have to BUILD a venture that IS AN ATTRACTIVE INVESTMENT. Ca or write -- I'll help if I can. Barry Stevens, PO Box 2747, Del Mar, CA 92014. 619-755-7231 ------------------------------ Date: 18 Oct 87 22:34:36 GMT From: violet.berkeley.edu!ed298-ak@jade.Berkeley.EDU (Edouard Lagache) Subject: Re: The Success of AI (Analysis of AI lack of progress). Anyone interested in the question of A.I. success (or lack of it) should have a look at Hubert Dreyfus's work. He has written two books which are critical of present A.I. methodologies, and make a purswasive argument for why present approaches to A.I. won't work. The books are: What Computers Can't Do; the Limits of Artificial Intelligence (Harper & Row, 1979) Mind over Machine; The Power of Human Intuition and Expertise in the Era of the Computer (co-authored with Stuart Dreyfus and Tom Athanasiou, The Free Press, 1986). It perhaps goes without saying that Hubert Dreyfus is one of the most disliked persons of A.I. researchers. However, no one in this field can really afford to not be aware of Dreyfus's concerns. Edouard Lagache School of Education U.C. Berkeley lagache@violet.berkeley.edu ------------------------------ Date: 17 Oct 87 22:09:05 GMT From: cbmvax!snark!eric@rutgers.edu (Eric S. Raymond) Subject: Re: The Success of AI In article <1922@gryphon.CTS.COM>, tsmith@gryphon.CTS.COM (Tim Smith) writes: > Computers do not process natural language very well, they cannot > translate between languages with acceptable accuracy, they > cannot prove significant, original mathematics theorems. I am in strong agreement with nearly everything else you say in this article, especially your emphasis on a need for a new paradigm of mind. But you are, I think, a little too dismissive of some real accomplishments of AI in at least one of these difficult areas. Doug Lenat's Amateur Mathematician program was a theorem prover equipped with a bunch of heuristics about what is 'mathematically interesting', essentially methods for grinding out interesting generalizations and combinations of known theorems. Lenat fed it the Zermelo-Frankel set theory axioms and let it run. After n hours of chugging through a lot of nontrivial but already-known mathematics, it 'conjectured' and then proved a bunch of new results on the number-theoretic properties of Pythagorean triples (3-tuples of integers of the form ). I was a theoretical mathematician at the time I saw the AM paper. It was *fascinating*. The program could probably have done a lot more, but it eventually choked on the size of its own LISP data structures. So at least one of your negative assertions is incorrect. I never heard of this line of research being followed up by anyone but Doug Lenat himself, and I've never been able to figure out why. He later wrote a program called EURISKO that (among other things) won that year's Trillion-Credit Squadron tournament (this is a space wargame related to the _Traveller_ role-playing game) and designed an ingenious fundamental component for VLSI logic. I think all this was in '82. > I believe the great success of AI has been in showing that > the old dualistic separation of mind and body is totally > inadequate to serve as a basis for an understanding of human > intelligence. Correct. But while recognizing this, let's not lose sight of the real accomplishments of AI in the purely-symbolic domain (whatever happened to Steve Harnad, anyhow?). I think AI has the same negative-definition problem that "natural philosophy" did when experimental science got off the ground -- that once people get a handle on some "AI" problem (like, say, playing master-level chess or automated proof of theorems) there's a tendency to say "oh, now we understand that; it's *just* computation, it's not really AI" and write it out of the field (it would be interesting to explore the hidden vitalist premises behind such thinking). So at any given time the referents for AI in peoples' minds are failures and unproved speculations, and the field goes through these manic-depressive cycles as it regroups around a new theory, problem or technology, explores it enough to make it useful for others, and then loses it to the rest of the world. Case in point: in the 1950s, *compilers* were considered "AI". I'm not old enough to remember that, but some of you may be. So, don't throw out the ship with the bath water -- er, that is, don't give up the baby -- er, oh, *you* know what I mean. AI is a useful category not in spite of all the ambiguity and confusion and excitement that surrounds it, but *because* of that. -- Eric S. Raymond UUCP: {{seismo,ihnp4,rutgers}!cbmvax,sdcrdcf!burdvax,vu-vlsi}!snark!eric Post: 22 South Warren Avenue, Malvern, PA 19355 Phone: (215)-296-5718 ------------------------------ End of AIList Digest ******************** 18-Oct-87 23:41:13-PDT,17628;000000000000 Mail-From: LAWS created at 18-Oct-87 23:34:55 Date: Sun 18 Oct 1987 23:29-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #239 - Neuromorphic Terminology, AI Successes, Logican Joke To: AIList@SRI.COM AIList Digest Monday, 19 Oct 1987 Volume 5 : Issue 239 Today's Topics: Neuromorphic Systems - Terminology, Opinion - The Success of AI, Humor - Two Logician Jokes, Philosophy - Flawed Human Minds ---------------------------------------------------------------------- Date: 15 Oct 87 14:16:55 GMT From: trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU (Stephen Smoliar) Subject: Re: Neural Networks - Pointers to good texts? In article <1465@ssc-vax.UUCP> dickey@ssc-vax.UUCP (Frederick J Dickey) writes: >In article <4191@well.UUCP>, wcalvin@well.UUCP (William Calvin) writes: >> We brain researchers sure get tired of hearing neural-like networks >> referred to as "neural networks", an established subject for 25 years since >> the days of Limulus lateral inhibition. > >I think the above says that "biological" neural nets have been studied as a >formal discipline for 25 years and that this great ancestry gives biology >prior claim to the term "neural nets". Assuming that this is a correct >interpretation, let me make the following observation. In 1943, McCulloch >and Pitts published a paper entitled "A logical calculus of the ideas >immanent in neural nets". Minsky and Papert (Perceptrons) state that this >paper presents the "prototypes of the linear threshold functions". This paper >stikes me as clearly being in the "neural net-like" tradition. Now >1987-1943 = 44. Also note that 44 > 25. Therefore, it apears that the >"neural net-like" guys have prior claim to the term "neural net". :-). Well . . . this is all rather silly. The PUBLISHED title of the classic paper by McCullogh and Pitts is "A Logigal Calculus of the Ideas Immanent in Nervous Activity." They NEVER use "neural net" as a technical term (or in any other capacity) in the paper. They ARE, however, concerned with a net model based on the interconnection of elements which they call neurons--appealing to properties of neurons which were known at the time they wrote the paper. Personally, I think Calvin has a point. Investigators who are searching the literature will probably benefit from cues which distinguish papers about actual physiological properties from those about computational models of those properties. ------------------------------ Date: 17 Oct 87 23:58:18 GMT From: ptsfa!well!wcalvin@ames.arpa (William Calvin) Subject: Re: Neural Networks - Pointers to good texts? I thank you all for the suggestions regarding renaming non-neural "Neural Networks" -- perhaps we can continue the discussion in the newsgroup comp.ai.neural-nets rather than here in comp.ai as such. William H. Calvin University of Washington NJ-15, Seattle WA 98195 [There is also the neuron%ti-csl.csnet@relay.cs.net list. -- KIL] ------------------------------ Date: 16 Oct 87 06:07:47 GMT From: ucsdhub!jack!man!crash!gryphon!tsmith@sdcsvax.ucsd.edu (Tim Smith) Subject: The Success of AI There is one humbling sense in which the work in AI in the past 20 or so years will help considerably in the ultimate understanding of human intelligence. If you look at concepts of the brain in the recent past, you see that whatever was the most current technological marvel served as a metaphor for the brain. In the early 20th century the brain was a telephone exchange. After WWII, the systems organization metaphor was often used (the brain was a large corporation, with a CEO, VPs, directors, etc.). It wasn't until computers came along that there was a metaphor for the brain powerful enough to be taken seriously. Once people started to try to imitate their brains on computers, some limitations became apparent. Interestingly enough, the limitations are not so much in the technological metaphor as in the present concept of the brain, or of the mind in general. There is no reason, in principle, that a very powerful digital computer cannot imitate a mind, *as long as a mind is some kind of abstract logic machine*. What AI has discovered (though it is very unwilling to admit it) is that this Cartesian (or even Platonic) concept of the mind is hopelessly inadequate as a basis for understanding human intelligence! To conceive of the human mind as a disembodied logic machine seemed like a great breakthrough to scientists and philosophers. If it was this, it could be studied and understood. If it wasn't this, then any scientific study of the mind (hence, of intelligence) appeared to be fruitless. The success rate in AI research (as well as most of cognitive science) in the past 20 years is not very encouraging. Predictions, based on very optimistic views of the problem domain, have not been met. A few successful spin-offs have occurred (expert systems, better programming tools and environments), but in general the history is one of failure. Computers do not process natural language very well, they cannot translate between languages with acceptable accuracy, they cannot prove significant, original mathematics theorems. What AI researchers and other cognitive scientists now have to face is fairly clear evidence that simulations of human intelligence, where human intelligence is modelled as a disembodied logic machine, are doomed to fail. Better hardware is not the solution. Connection machines or simple silicon neural nets are not the answer. A better concept of "mind" is what is needed now. This is not to say that AI research should halt, or that computers are not useful in studying human intelligence. (They are indispensable.) What I think it does mean is that one or more really original theoretical paradigms will have to be developed to begin to address the problems. One possible source of a new way of thinking about the problems of modelling human intelligence might be found in a revolution that is beginning in the cognitive sciences. This revolution is of course not accepted by most cognitive scientists; many are not even aware of it. It is difficult to characterize the revolution, but it essentially rejects the Cartesian dualism of mind and body, and recognizes that an adequate description of human intelligence must take into account aspects of human physiology, experience, and belief that cannot *now* be modelled by simple logic (e.g., programs). For one example of this new way of thinking, see the recent book by the linguist George Lakoff, entitled "Women, Fire, and Dangerous Things." (Neither the book nor the title are frivolous.) I believe the great success of AI has been in showing that the old dualistic separation of mind and body is totally inadequate to serve as a basis for an understanding of human intelligence. -- Tim Smith INTERNET: tsmith@gryphon.CTS.COM UUCP: {hplabs!hp-sdd, sdcsvax, ihnp4, ....}!crash!gryphon!tsmith UUCP: {philabs, trwrb}!cadovax!gryphon!tsmith ------------------------------ Date: 18 Oct 87 01:39:46 GMT From: PT.CS.CMU.EDU!SPICE.CS.CMU.EDU!spe@cs.rochester.edu (Sean Engelson) Subject: Re: The Success of AI Given a sufficiently powerful computer, I could, in theory, simulate the human body and brain to any desired degree of accuracy. This gedanken-experiment is the one which put the lie to the biological anti-functionalists, as, if I can simulate the body in a computer, the computer is a sufficiently powerful model of computation to model the mind. I know, for example, that serial computers are inherently as powerful computationally as parallel computers, though not as efficient, as I can simulate parallel processing on essentially serial machines. So we see, that if the assumption that the mind is an inherent property of the body is accepted, we must also accept that a computer can have a mind, if only by the inefficient expedient of simulating a body containing a mind. -Sean- -- Sean Philip Engelson I have no opinions. Carnegie-Mellon University Therefore my employer is mine. Computer Science Department ---------------------------------------------------------------------- ARPA: spe@spice.cs.cmu.edu UUCP: {harvard | seismo | ucbvax}!spice.cs.cmu.edu!spe ------------------------------ Date: Thu 15 Oct 87 16:36:24-CDT From: David Throop Subject: Two Logician Jokes #G-120-97A I was hanging out at the Logicians Union Hall the other day and the place was full of logicians, poring over logician's manuals and exchanging gossip. Well, every so often, one of them would call out a number, and all of the others would laugh real hard. Then they'd all go back to whatever they were doing. This seemed real odd behavior for such logical people. So I asked Robert, who's a logician friend of mine there, what was going on. "Hey, this is a hall for logicians," he said. "A while back, we collected all of the jokes that we could prove were funny and put them in a catalog. Everybody here's read it. Now when somebody wants to tell a joke, they just call out its serial number." And he showed me the logical joke catalog. I thumbed through it for a while. Found a joke I liked. And at an opportune time, I called it out: "G-120-97B!" Nobody laughed. I turned to Robert and said "So how come they didn't laugh?" He shrugged. "You didn't tell it right." ============================================================================= G-120-97C I was hanging out at the Logicians Union Hall the other day and the place was full of logicians, poring over logician's manuals and exchanging gossip. Well, every so often, one of them would call out a number, and all of the others would laugh real hard. Then they'd all go back to whatever they were doing. This seemed real odd behavior for such logical people. So I asked Robert, who's a logician friend of mine there, what was going on. "Hey, this is a hall for logicians," he said. "A while back, we collected all of the jokes that we could prove were funny and put them in a catalog. Everybody here's read it. Now when somebody wants to tell a joke, they just call out its serial number." And he showed me the logical joke catalog. I thumbed through it for a while. Found a joke I liked. Actually, THIS was the joke. This joke I'm telling you right now, it's numbered G-120-97C. And here's where it gets hard. Because if the joke is funny, then the logicians laugh, and that spoils the punchline. And the joke isn't funny any more. But if the logicians will laugh at any funny joke. So if they don't laugh, it's because the joke isn't funny. But then the punchline works and its funny again. So I can't tell you whether or not the logicians laughed. Either way, it spoils the punchline. ------------------------------ Date: 16 Oct 87 12:56:21 GMT From: ihnp4!homxb!genesis!odyssey!gls@ucbvax.Berkeley.EDU (g.l.sicherman) Subject: Re: The Job Hunt > > > Mr X. goes to an employment interview and gets angry or flustered and > > > says something that causes him to be rejected. Without knowing how his > > > mind works you can conclude it was flawed. > > > > And you could be wrong. Most likely Mr. X. didn't want the job after > > all. He only wanted you to think he wanted the job. Give him credit > > for some intelligence! > > Also flawed from Mr. X's point of view. Sicherman argues that X only > seemed to get angry or flustered, in order to make sure the company > didn't make him an offer, because during the interview he decided he > didn't want a job with them. If I attributed Mr. X's actions to > intelligence I would expect him to conclude gracefully, let them make > an offer, and reject the offer, without making a bad impression on > somebody who later might be in a position to offer him a job in another > company. And I don't care whether you blame emotions or habits. You misunderstood me. I suggested not that X *seemed* to get angry, but that he genuinely got angry. Emotions are not some kind of side effect-- they serve a constructive purpose. Anger, in particular, drives away or destroys things that threaten your well-being. Most likely Mr. X wants to avoid getting a job, but wants people in general or certain people in particular to think he wants a job. It happens all the time! You're wasting your time when you pontificate to Mr. X. He's not going to tell a back-seat driver like you what he really wants. > > By this criterion, we are all flawed. > That's exactly what I meant. Well, it's a useless and insulting criterion. -- Col. G. L. Sicherman ...!ihnp4!odyssey!gls ------------------------------ Date: 16 Oct 87 17:07:02 GMT From: ihnp4!homxb!houdi!marty1@ucbvax.Berkeley.EDU (M.BRILLIANT) Subject: Re: The Job Hunt In article <333@odyssey.ATT.COM>, gls@odyssey.ATT.COM (g.l.sicherman) writes: > > You misunderstood me. I suggested not that X *seemed* to get angry, but > that he genuinely got angry. Emotions are not some kind of side effect-- > they serve a constructive purpose. Anger, in particular, drives away > or destroys things that threaten your well-being. > > Most likely Mr. X wants to avoid getting a job, but wants people in > general or certain people in particular to think he wants a job. It > happens all the time! You're wasting your time when you pontificate > to Mr. X. He's not going to tell a back-seat driver like you what he > really wants. Do we need a definition of anger? Anger, as I understand it, is an emotion that catalyzes physical actions but interferes with reason. I agree that Mr. X may rationalize his action, but I don't believe it was his best choice. > > > By this criterion, we are all flawed. > > > That's exactly what I meant. > > Well, it's a useless and insulting criterion. Pardon me. I thought what we all needed was a little humility. If Col. G. L. Sicherman thinks either that he is perfect, or that I am perfect, I disagree. Tentatively. In my simplistic view, the mind is a complex system that came to be what it is through variation and natural selection. It has functions that we don't understand, adaptations for purposes we don't understand, and adaptations for purposes that no longer exist. If it's perfect, that's a marvelous coincidence. If the aim of artificial intelligence is to model the human mind, Col. Sicherman and I seem to agree that it's not enough. To model anger, for instance, we also need artificial emotion. But if the aim of artificial intelligence is to create a purely intelligent entity without maladaptive emotions, Col. Sicherman and I would disagree. I believe that at least some emotional responses are maladaptive and would not exist in a perfect intelligence, while he apparently believes the human mind is perfect and cannot be improved upon. So let us agree to disagree, and, as I suggested in an earlier article, let some AI researchers model the human mind, while others build something better adapted to specific tasks. M. B. Brilliant Marty AT&T-BL HO 3D-520 (201)-949-1858 Holmdel, NJ 07733 ihnp4!houdi!marty1 ------------------------------ Date: 17 Oct 87 08:52:56 GMT From: imagen!atari!portal!cup.portal.com!tony_mak_makonnen@ucbvax.Berk eley.EDU Subject: Re: Flawed human minds There is strange and profound truth to the following statements All the universe is the brain All you know is the mind The second statement is more daring than the first . It is necessitated by the need to posit something that is more than the physical parts of brain . Assume a completely isolated , closed system capable of reflection . I submit that such a thinking thing could not posit a essential flaw in its make up .We see here many individual manifestations of mind talking about flaws that one can only assume must be attributed to the brain .What is that which stands back and reflects on the flawed function of that very instrument without which it would be a "null"in this universe ? Can we call it "I" or "mind" . But then some seem to posit other "I"s than can stand back and look at the first "Iand so on . Very confusing once we leave the safety of behavioral psych . What seems to the morale at this point ? Accept the obvious fact that the brain is not very efficient at calculative functions , and the equally true fact that it is capable of creating machines That can do that much better . Forget about the other abstract stuff . This mind knows the limts of some brain functions and compensates for them . It has so far proved adequate for the primary directive "the survival of the species and life " . I submit to the members of this jury that we cannot yet say that it is flawed . However should we reach the ultimate folly of self destruction then only the absence of an audience and judge will prevent a definitive verdict . ------------------------------ End of AIList Digest ******************** 21-Oct-87 22:43:07-PDT,15922;000000000001 Mail-From: LAWS created at 21-Oct-87 22:30:22 Date: Wed 21 Oct 1987 22:26-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #240 - Net Mail to UK, Lisp Books, Logician Jokes To: AIList@SRI.COM AIList Digest Thursday, 22 Oct 1987 Volume 5 : Issue 240 Today's Topics: Queries - Neuromorphic Systems Sources & Cash Flow and Expert Systems & LISP on the AMIGA & Explanations in XPS, Bindings - Net Mail to the UK, Reviews - Introductory Books on Lisp, Humor - Another Numbered Joke Joke ---------------------------------------------------------------------- Date: Tue 20 Oct 87 21:07:39-EDT From: John C. Akbari Subject: neuro sources anyone have the source code for either of the following? kosko, bart. constructing an associative memory. _byte_ sept. 1987 jones, w.p. & hoskins, j. back-propagation. _byte_ oct. 1987. any help would be appreciated. John C. Akbari PaperNet 380 Riverside Drive, No. 7D New York, New York 10025 USA SoundNet 212.662.2476 (EST) ARPANET & Internet akbari@CS.COLUMBIA.EDU BITnet akbari%CS.COLUMBIA.EDU@WISCVM.WISC.EDU UUCP columbia!cs.columbia.edu!akbari ------------------------------ Date: Mon, 19 Oct 87 20:09:29 GMT From: A385%EMDUCM11.BITNET@wiscvm.wisc.edu Subject: Literature on Cash-flow & Expert Systems Date: 19 October 1987, 20:07:13 GMT From: Javier Lopez Torres Tf: (91) 7113887 A385 at EMDUCM11 C/ Mirlo 1. 28024 Madrid -Spain- To: AILIST-R at SRI Hello AI Community from Spain!!! We have just begun to developpe an expert system for cash-flow in Common-Lisp, but we'd like first to acquire some theoretical background on this subject. So please, could anyone of you suggest any good text about expert systems and cash-flow??. Thank you very much in advance for any help or suggestion. Yours Javier Lopez UNiversidad Complutense de Madrid ------------------------------ Date: 21 Oct 87 13:30:44 GMT From: oliveb!amiga!cbmvax!phillip@ames.arpa (Phillip Lindsay GUEST) Subject: LISP on the AMIGA. [Eat|Me] I would like to hear from people working on anything related to LISP and/or AI on the Amiga. This is important since I am trying to solicit a port of a LISP product. Any general interest also welcome. (the more bullets the better) ------------------------------ Date: 22 Oct 87 01:06:41 GMT From: spieker@uklirb.UUCP Subject: Explanations in XPS - (nf) Article-I.D.: uklirb.40000002 Hi, can anybody outthere send me an (extended) bibliography on the subjects of - Explanation Generation in Expert Systems - User Modelling Thanks Peter Spieker Universitaet Kaiserslautern Fachbereich Informatik P.O.Box 3049 D-6750 Kaiserslautern FRG UUCP: ...mcvax!unido!uklirb!spieker ------------------------------ Date: Tue, 20 Oct 87 00:30:58 EDT From: brant@linc.cis.upenn.edu (Brant Cheikes) Subject: net mail to the UK If you happen to know the Usenet name of a host in the UK, then as a temporary solution, you can use "ukhost!ukuserid@uunet.uu.net". The ARPAnet host uunet.uu.net is an (official?) arpa/usenet gateway and knows how to route mail to all known uucp hosts, including those in the UK. I, for example, have been corresponding with my advisor, Bonnie Webber, who's on sabbatical at Edinburgh, by addressing mail to "eusip!bonnie@uunet.uu.net". Unfortunately, I think this only works for Unix hosts on usenet. Mail to people at ucl-cs can be sent to "ucl-cs!user@uunet.uu.net". Hope that helps. Brant Brant Cheikes University of Pennsylvania ARPA: brant@linc.cis.upenn.edu Computer and Information Science ------------------------------ Date: 19 Oct 87 11:56 PDT From: hayes.pa@Xerox.COM Subject: email to UK Some news about UK email. The UCL gateway has recently introduced a policy ( see official notice reproduced below ) which polices traffic through the gateway to an alarming extent. I have talked ( well, listened ) to some moderately senior UK administrators about this and have been told that this is being forced on them by the US military, but you know what politicians are. Anyway, the effect has been to have hackers of unknown competence start fiddling with a working system, with predictable results. There seems to be considerable confusion: the UCL locals ( email to liaison@cs.ucl.ac.uk ) insist that mail into the UK should go through regardless of source and that only outgoing mail will be policed, but the official bulletin says otherwise. In the interim there is even more confusion here: The new `correct' address, say the UCL hackers, is nss.cs.ucl.ac.uk, but the host tables as of a couple of weeks ago had it as synonymous with the old cs.ucl.ac.uk. ( and, by the way, with UCL-CS ) . The UCL people were horrified when I told them of this so maybe things have changed ( cf KILs reply to Yorick ), but for a while the following hack, suggested by Doug Faunt at Schlumberger, worked just fine: mail to whoever%nss.cs.ucl.ac.uk@cs.ucl.ac.uk This apparently made the ucl machine forward to itself and then believe the nss prefix. I am testing this again now but dont have results yet. The NS2 having a different socket number is very encouraging, I bet this is the missing NSS with a typo: I am testing this as well. Pat Hayes PS. Let me suggest that all users of netmail to the UK send their comments on the following to whoever they think might be inclined to listen, such as a senator or M.P. ----------------- From: liaison@NSS.Cs.Ucl.AC.UK Subject: Authorisation Information Sender: daemon@NSS.Cs.Ucl.AC.UK To: Witty.pa TO UNAUTHORISED USERS OF THE UCL GATEWAY SERVICE PROJECT: Access control has been introduced to the UCL ARPA/Janet Gateway, so that only authorised users of the Service may send traffic through the Gateway. This is because of restrictions imposed on the Service by its funding bodies. If you wish to exchange mail with users on the other side of the Gateway, an application must be made to gain authorisation. It is most appropriate for the UK user to apply. Mail is authorised by either sender or receiver, so that a US user is able to send mail to an authorised UK mailbox. If you are a US user, please contact your UK colleague by some other means, to explain what is now happening - he/she may be unaware of these developments. This applies to all the Internet networks reached from UCL via Arpanet (including Usenet mail to or from US hosts that is routed via UCL), and to PSS in the UK. The actual authorisation mechanism depends on the registration of mailboxes belonging to the applicant - the program necessary to do this is available when and if authorisation is given. One application is made per project group, in the name of the principal investigator of the project. The mailboxes of all the members of the group can then be registered as being associated with the authorised user. ________________________________________________________________ Further information can be obtained from auto-mailboxes. 1. For an application form (for UK users) and Introductory document to the Gateway Service send a message to the auto-mailbox: application-form@ucl-cs 2. Some JANET sites have a contact who is willing to assist with mail registration problems. A list can be obtained from the auto-mailbox: local-help@ucl-cs 3. A general bulletin board for users of the UCL Gateway can be obtained similarly from the auto-mailbox: netnews@ucl-cs. 4. The 'mreg' program allows authorised users to register mailboxes for which they are responsible. A guide to using the 'mreg' program can be obtained from the auto-malbox: mreg-help@ucl-cs No text should be included with messages to auto-mailboxes. ------------------------------ Date: Mon, 19 Oct 87 08:58:36 EDT From: rapaport@cs.Buffalo.EDU (William J. Rapaport) Subject: lisp books An excellent self-study book on Lisp is: Shapiro, Stuart C., LISP: An Interactive Approach (Computer Science Press) It's dialect-independent, and assumes that the reader is sitting in front of a terminal running Lisp while reading the book. William J. Rapaport Assistant Professor Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3181 uucp: ..!{ames,boulder,decvax,rutgers}!sunybcs!rapaport internet: rapaport@cs.buffalo.edu [if that fails, try: rapaport%cs.buffalo.edu@relay.cs.net or: rapaport@buffalo.csnet ] bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: 19 Oct 87 10:54:01 edt From: Walter Hamscher Subject: Introductory books on Lisp Date: Fri, 16 Oct 87 11:03:46 PDT From: glasgow@marlin.nosc.mil (Michael G. Glasgow) I am new to AIList and AI programming and want to learn Lisp. I have been looking through Steele's book, Common Lisp", and have discovered that this is more of a reference manual than a beginners guide. What I am wondering is if anyone can give me the names of some good introductory Lisp books to get me started. There are several. Here are two: Winston, Horn, "LISP". Addison-Wesley (1984 I think). Teaches you common lisp from the atoms on up. Charniak, Riesbeck, McDermott "Artificial Intelligence Programming" Lawrence Erlbaum (1980). What every AI programmer should know, though unfortunately the lisp dialect is getting a bit dated. Two others I know of but have never had the opportunity to use: Wilensky, "Common LISPcraft". Norton, 1984. Brooks, "Programming in Common Lisp." MIT Press, 1985. You will undoubtedly hear from the partisans of other books. ------------------------------ Date: 20 Oct 87 02:55:25 GMT From: voder!apple!andyr@decwrl.dec.com (Andy Rundquist) Subject: Re: Introductory books on Lisp In article <8710161803.AA06962@marlin.nosc.mil>, glasgow@MARLIN.NOSC.MIL (Michael G. Glasgow) writes: > > > I am new to AIList and AI programming and want to learn Lisp. > I have been looking through Steele's book, Common Lisp", and > have discovered that this is more of a reference manual than a > beginners guide. What I am wondering is if anyone can give me > the names of some good introductory Lisp books to get me started. > > Thanks in Advance, > > michael To me, the best (and most enjoyable) Lisp introduction can be found in: _The Little Lisper_ by D. Freidman. Andy (Now CONS a piece of cake into your mouth) ------------------------------ Date: Tue, 20 Oct 87 11:27:25 PDT From: Stephen Smoliar Subject: Re: Introductory books on Lisp Back in the dark ages when I was teaching LISP, I used to rely heavily on THE LITTLE LISPER by Daniel Friedman. I felt that the important thing about learning LISP was getting comfortable with expressing yourself in a functional style and using the format of recursive definitions. Friedman does an excellent job of walking you through a broad variety of examples. You emerge from this book with a good sense of the power of a "pure" applicative style of LISP programming. Having done so, you are now ready for the "real world" provided by the particular dialect of LISP you will actually be using. ------------------------------ Date: 20 Oct 87 12:45:48 GMT From: kddlab!secisl.seclab.junet!tau@uunet.UU.NET ("Yatchan" TAUCHI) Subject: Re: Introductory books on Lisp In article <8710161803.AA06962@marlin.nosc.mil>, glasgow@MARLIN.NOSC.MIL (Michael G. Glasgow) writes: > I have been looking through Steele's book, Common Lisp", and > have discovered that this is more of a reference manual than a > beginners guide. It's not a good text book to Lisp beginners, but just specification of COMMON- LISP. > What I am wondering is if anyone can give me > the names of some good introductory Lisp books to get me started. I think there are not many good books on CommonLisp yet. I recommend "Common LISPcraft" by Robert Wilensky, Norton $26.95. His book, "LISPcraft" was very good text book on FranzLisp. I think it's easy to understand how to write CommonLisp program. ---- Yasuyuki TAUCHI, SECOM IS-Lab, Tokyo JAPAN NET: tau%seclab.junet@uunet.UU.NET UUCP: ...!seismo!kddlab!titcca!secisl!tau ------------------------------ Date: Mon 19 Oct 1987 08:47 CDT From: UUCJEFF%ECNCDC.BITNET@wiscvm.wisc.edu Subject: >Everybody here's read it. Now when somebody wants to tell a joke, th >just call out its serial number." And he showed me the logical joke >catalog. >I thumbed through it for a while. Found a joke I liked. And at an >opportune time, I called it out: "G-120-97B!" >Nobody laughed. G-120-97B Thats an IBM manual right? ------------------------------ Date: Mon, 19 Oct 87 11:14:16 BST From: Graham Higgins Subject: Two Logician Jokes That's a shame. Why does there have to be a concrete catalogue? It prohibits a variation which is arguably funnier ... can I change it around bit ?? .... I was hanging out at the Logicians Union Hall the other day and the place was full of logicians, poring over logician's manuals and exchanging gossip. Well, every so often, one of them would call out a number, and all of the others would laugh real hard. Then they'd all go back to whatever they were doing. This seemed real odd behavior for such logical people. So I asked Robert, who's a logician friend of mine there, what was going on. "Hey, this is a hall for logicians," he said. "A while back, we collected all of the jokes that we could prove were funny and allocated each one a number. Everybody here knows them. Now when somebody wants to tell a joke, they just call out its number." I thought about this for a while. Then I asked Robert if I could tell a joke, you know, try out the system. He said it would be OK by him, so I called out a number: "1209!" Nobody laughed. I turned to Robert and said "So how come they didn't laugh?" He shrugged. "You didn't tell it right." I asked for another try. Once more, Robert said it would be OK by him. I called out a different number: "83417!". Everybody collapsed in fits of mirth. I turned to Robert, feeling pleased. "Told _that_ one OK, didn't I?", I said. Robert was nearly helpless with laughter. He gasped, in between guffaws, "Haven't heard that one before". ------------------------------ Date: Tue, 20 Oct 87 11:22:42 PDT From: Stephen Smoliar Subject: Another numbered joke joke. [Same first three paragraphs.] One of the logicians called out: "G-120-97D!" Suddenly, one logician had a fit of uncontrollable laughter; and it took some time before he calmed down. I turned to Robert and asked, "What happened?" He replied, "Probably that logician had never heard that joke before." Post Script: I originally heard these jokes in the context of Hollywood producers exchanging jokes at lunch (even before I moved out to the shadow of Hollywood). ------------------------------ End of AIList Digest ******************** 21-Oct-87 22:56:35-PDT,9198;000000000000 Mail-From: LAWS created at 21-Oct-87 22:47:58 Date: Wed 21 Oct 1987 22:46-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #241 - Seminars, Course in Information Processing To: AIList@SRI.COM AIList Digest Thursday, 22 Oct 1987 Volume 5 : Issue 241 Today's Topics: Seminars - Crystallizing Theories out of Knowledge Soup (SU) & Event-Based Reasoning for Multiagent Domains (Bendix & BBN) & Computing in the Year 2001 (Aston) & Restricted And-Parallelism for Logic Programs (SMU), Course - Information Processing ---------------------------------------------------------------------- Date: Tue 20 Oct 87 17:47:25-PDT From: Marcelo Hoffmann Subject: Seminar - Crystallizing Theories out of Knowledge Soup (SU) John Sowa, a member of the IBM Systems Research Institute will be giving a talk titled "Crystallizing Theories out of Knowledge Soup (knowledge base)", on Thursday, October 22, at 7:00 PM in Room 380C Mathematics Department, Stanford University (while facing the Quad from Palm Drive, in the nearest, right hand corner of the Quad). The talks is sponsored by the IEEE Computer Society. Abstract: "The most challenging problems for AI arise from the difficulty of characterizing the knowledge soup, analyzing it, and codifying it in formal symbolic terms. These problems appear in many different guises in knowledge acquisition, machine learning, metaphor analysis, nonmonotonic reasoning, and reasoning with uncertainty. No complete, formal solutions are possible, but methods of conceptual analysis, belief revision, and dynamic type hierarchies permit special-case subtheories to be crystallized out of the knowledge soup as needed. This talk will use conceptual graphs as the formalism for representing the crystallized theories and show how they can be used with belief revision systems and dynamically changing type hierarchies". Attendance is free. ------------------------------ Date: Thu, 15 Oct 87 08:49 EDT From: DON%atc.bendix.com@RELAY.CS.NET Subject: Seminar - Event-Based Reasoning for Multiagent Domains (Bendix & BBN) Where: Allied-Bendix Aerospace Technology Center 9140 Old Annapolis Rd (MD 108) Columbia, MD 21045 When: 28 October 1987, 1:30pm Who: Amy L. Lansky SRI International, Artificial Intelligence Center What: Localized Event-Based Reasoning for Multiagent Domains This talk will present GEM, a structured, event-based concurrency model, and GEMPLAN, a multiagent planner based on this model. A key focus of this work has been the development of localized techniques for domain representation and reasoning. Such techniques partition domain descriptions and reasoning tasks according to the regions of activity within a domain. GEM's use of locality is beneficial for alleviating the frame problem in multiagent domains. GEMPLAN is a planning architecture based on localized planning search spaces. By explicitly utilizing constraint and property localization, GEMPLAN can pinpoint and rectify interactions among regional search spaces, thereby reducing the burden of ``interaction analysis'' ubiquitous to most planning systems. Directions and RSVP (optional but helpful for planning): Roz Alme (301) 964-4106 or ROZ@ATC.BENDIX.COM. Marc Vilain reports that the same seminar will be given at BBN: 10 Moulton Street 2nd floor large conference room 10:30 am, Monday October 26 ------------------------------ Date: 16-OCT-1987 16:55:10 From: HANCOXPJ@MAIL.ASTON.AC.UK Subject: Seminar - Computing in the Year 2001 (Aston) From: Dr P J Hancox Dept: Computer Science Tel No: 021 359 3611 X4652 Aston University Department of Computer Science and Applied Mathematics Seminar Wednesday 28 October 1987 at 3.00 pm in Room 550, Main Building Computing in the year 2001 Brian Oakley Director, The Alvey Directorate, London The key to the advance in computing over the last 20 years has been the inexorable increase in the speed, power and memory capacity of the silicon chip. Will this continue and, if so, for how long? The talk will consider the performance of the integrated circuit in the year 2001, and the resulting power of the processor on a chip. Well before the turn of the century multi-processors will have become common place, so that system power will far exceed the individual processor power. And what will this power be used to do? The talk will end by considering the new applications, particularly the spread of so-called AI applications such as Expert Systems, Natural Language, Voice and Image Processing. Chairman: Dr B Gay. Enquiries: JANET: compsci@uk.ac.aston.mail uucp: seismo!mcvax!ukc!aston!compsci Computer Science, Aston University, Birmingham, B4 7ET, United Kingdom + 44 21 359 3611 extn 5313 ------------------------------ Date: Mon, 19 Oct 1987 00:32 CST From: Leff (Southern Methodist University) Subject: Seminar - Restricted And-Parallelism for Logic Programs (SMU) Restricted And-Parallelism for Logic Programs SPEAKER: Doug DeGroot LOCATION: 315 SIC Southern Methodist University Texas Instruments Inc. TIME: 1:30 pm ABSTRACT A number of parallel execution models have been proposed for the highly-parallel execution of logic programs. Most of these center on various forms of and-parallelism and/or or-parallelism. While the majority of work in Europe and Japan seems to have focused on or-parallel models, research in the United States had focused more on and-parallelism. Some of the reasons for this will be examined, and a number of models for and-parallelism and related research will be mentioned. Then a specific model, called Restricted And-Parallelism (RAP) will be described in detail. The discussion will focus on the model itself, techniques for the automatic compilation of Prolog programs into RAP graph expressions, and the proper handling of side-effects in a parallel execution environment, and global data-dependency analysis for better program decomposition. An overview of RAP-related research efforts in other parts of the world will also be discussed. Finally, topics for future research will be discussed as time permits. ------------------------------ Date: 16 Oct 87 10:52:40 GMT From: mcvax!cernvax!cui!pun@uunet.uu.net (PUN Thierry) Subject: Course - Information Processing (I am forwarding the following annoucement; please enquire directly to the address below. TP.) The Swiss Federal Institute of Technology, Lausanne presents a graduated program leading to a M.S. in information processing with applications to systems signals and images. The closing of registrations is November 15 1987. This program starting in March 1988 consisting of two terms study (one year) including lectures, exercices and workshops is based on the following themes : A) Communication system theory (160 hours) A1) Systems theory (30 hours) A2) Information theory (70 hours) A3) Detection and estimation (60 hours) B) Digital signal and image processing B1) Digital signal processing (80 hours) B2) Digital image processing (90 hours) C) Pattern recognition and scene analysis (150 hours) C1) Pattern recognition (90 hours) C2) Scene analysis (60 hours) D) Real time information processing (145 hours) D1) Speech processing (60 hours) D2) Signal processing and VLSI architecture (85 hours) These lectures, exercices and workshops will be followed by a research project during 6 months in 1989. The lecturers are : M. Kunt (course director), F. Ade, M. Bellanger, D. Bonvin, J. Caelen, G. Caelen-Haumont, G. Coray, F. de Coulon, P. Dewilde, O. Faugeras, W. Fichtner, G. Granlund, C. Gueguen, B. Guerin, M. Hasler, J.P. Haton, H. Hugli, R. Ingold, O. Kubler, R. Longchamp, H. Nussbaumer and Ch. Sorin. To get more informations please contact the secretariat du Laboratoire de traitement des signaux de l'EPFL, Departement d'electricite, EPFL Ecublens, CH 1015 Lausanne, SWITZERLAND, Tel. (4121) 472624 or 472601, Telex 454062 EPFVD CH, Telefax (4121) 474660 PUN CGEUGE51 10/14/87 THIERRY PUN cvnet@yorkvm1 10/14/87 For posting on the net (start ------------------------------ End of AIList Digest ******************** 21-Oct-87 23:06:30-PDT,21082;000000000000 Mail-From: LAWS created at 21-Oct-87 22:57:24 Date: Wed 21 Oct 1987 22:53-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #242 - Successes of AI, Automated Discovery To: AIList@SRI.COM AIList Digest Thursday, 22 Oct 1987 Volume 5 : Issue 242 Today's Topics: Comments - The Success of AI Representation - Lenat's AM Program ---------------------------------------------------------------------- Date: 19 Oct 87 09:54:22 edt From: Walter Hamscher Subject: The Success of AI Date: 18 Oct 87 01:39:46 GMT From: PT.CS.CMU.EDU!SPICE.CS.CMU.EDU!spe@cs.rochester.edu (Sean Engelson) Given a sufficiently powerful computer, I could, in theory, simulate the human body and brain to any desired degree of accuracy. * * * Don't forget to provide all the sensory input provided by being in, moving around in, and affecting the world. Otherwise you'll be simulating a catatonic. Do the terminally catatonic have minds? ------------------------------ Date: 19 Oct 87 10:27:22 edt From: Walter Hamscher Subject: The Success of AI Date: 17 Oct 87 22:09:05 GMT From: cbmvax!snark!eric@rutgers.edu (Eric S. Raymond) * * * I never heard of this line of research being followed up by anyone but Doug Lenat himself, and I've never been able to figure out why. He later wrote a program called EURISKO that (among other things) won that year's Trillion-Credit Squadron tournament (this is a space wargame related to the _Traveller_ role-playing game) and designed an ingenious fundamental component for VLSI logic. I think all this was in '82. See Lenat & J.S. Brown in AI Journal volume 23 #3, 1984: "Why AM and EURISKO Appear to Work". The punchline of the article (briefly) is that AM seems to have succeeded in elementary set theory because its own representation structures (i.e., lists), were particularly well suited to reasoning about sets. It started breaking down at exactly the places where its representation was inadequate for the concepts. For example, there was no obvious way to move from its representation of the number n as a list of length n, to a positional representation that would make it more likely to discover things like logarithms. Furthermore, its operations on procedures involved local modifications to procedures expressed as list structures, and as long as the procedures were compact these "mutations" were likely to produce interesting new behavior, but as the procedures get more complex, arbitrary random local modifications had a vanishingly low success ratio. Hence it would seem that direction to go from this insight is to make programs that can learn new representations. There are probably not enough people working on that. But anyway this is getting off the subject, which is whether AI has had any successes. Whether you want to count AM as a success is half-empty / half-full issue; the field surely learned something from it, but it surely hasn't learned enough. ------------------------------ Date: 19 Oct 87 17:47:45 GMT From: brian@sally.utexas.edu (Brian H. Powell) Subject: Re: The Success of AI In article <228@snark.UUCP>, eric@snark.UUCP (Eric S. Raymond) writes: > Doug Lenat's Amateur Mathematician program was a theorem prover equipped with > a bunch of heuristics about what is 'mathematically interesting', > [...] > > After n hours of chugging through a lot of nontrivial but already-known > mathematics, it 'conjectured' and then proved a bunch of new results on the > [...] I feel compelled to challenge this, but not necessarily the rest of your article. AM wasn't a theorem prover. From the July, 1976 dissertation: 7.2.2 Current Limitations [...] AM has no notion of proof, proof techniques, formal validity, heuristics for finding counterexamples, etc. Thus it never really establishes any conjecture formally. ---end of excerpt--- The dissertation goes on to briefly suggest ways of adding this capability, but as I understand it, no one ever has. Lenat himself, as I recall, thought it was more interesting to do more work towards heuristics than proving. EURISKO was the result of that. (i.e., you might get more power if you could spend part of your time conjecturing heuristics in addition to conjecturing about particular problems.) AM is a neat program, but by many views it's overrated. It's great that it conjectures all these neat theorems, but my impression is that it does quite a bit of floundering to find them. I think it also spends a lot of time floundering without finding anything useful, also. (A program run isn't guaranteed to think of something clever.) Finally, it's not clear that the program is really intelligent enough to realize that it's just conjectured something intelligent. (I would bet that there are a lot of things AM has considered uninteresting that humans consider interesting, and vice-versa.) A human can monitor AM and modify the priority of certain tasks if the human feels AM is studying the wrong thing. A human is practically required for this purpose if AM is to do something especially clever. This turns AM more into a search tool than an autonomous program, and I don't think a tool is what Lenat had in mind. If you read the summaries of AM, you think it's powerful. Once you read the entire dissertation, you realize it's not quite as great a program as you had thought, but you still think it's good research. Brian H. Powell UUCP: ...!uunet!ut-sally!brian ARPA: brian@sally.UTEXAS.EDU ------------------------------ Date: 20 Oct 87 06:30:06 GMT From: mcvax!cernvax!ethz!srp@uunet.uu.net (Scott Presnell) Subject: Re: The Success of AI In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: > >Given a sufficiently powerful computer, I could, in theory, simulate >the human body and brain to any desired degree of accuracy. This Horse shit. The problem is you don't even know exactly what you are simulating! I suppose you could say it's all a problem of definition, however even with your assumtion that the mind is a integral part of the body I still claim that you don't know what you're simulating. For instance, dreams, are they logical?, do they fall in a pattern?, a computer has got to have them to be a real simulation of a body/mind, but you cannot simulate what you cannot accurately describe. Let's get down to a specific case: I propose that given any amount of computing power, you could not presently, and probably will never be able to simulate me: Scott R. Presnell. My wife can be the judge. This may sound reactionary, that's because that's the way I responded internally to this first statement. I apologize if I've jumped into a discussion too quickly, I don't have time to read the previous discussions right now. Scott Presnell Organic Chemistry Swiss Federal Institute of Technology (ETH-Zentrum) CH-8092 Zurich, Switzerland. uucp:seismo!mcvax!cernvax!ethz!srp (srp@ethz.uucp); bitnet:Benner@CZHETH5A ------------------------------ Date: 21 Oct 87 05:30:58 GMT From: ucsdhub!jack!man!crash!gryphon!tsmith@sdcsvax.ucsd.edu (Tim Smith) Subject: Re: The Success of AI In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: +===== | Given a sufficiently powerful computer, I could, in theory, simulate | the human body and brain to any desired degree of accuracy. This | gedanken-experiment is the one which put the lie to the biological | anti-functionalists, as, if I can simulate the body in a computer, the | computer is a sufficiently powerful model of computation to model the | mind. I know, for example, that serial computers are inherently as | powerful computationally as parallel computers, though not as | efficient, as I can simulate parallel processing on essentially serial | machines. So we see, that if the assumption that the mind is an | inherent property of the body is accepted, we must also accept that a | computer can have a mind, if only by the inefficient expedient of | simulating a body containing a mind. | -Sean- +===== My claim is, specifically, that you cannot simulate a human being (body and mind) with a digital computer, either in theory or practice. Not a few people with whom I am in basic agreement would claim that, well, it just *might* be conceivable in theory, but you could never do it in practice. I'ts not clear what is meant by "in theory" here. It sounds like an unacceptable hedge. You might, for example, claim that with a very large number of computers, all just at the edge of the speed boundaries dictated by the laws of physics in the most advanced materials imaginable, you could simulate a human body and mind--but not in real time. But the simulation would have to be in real time, because humans live in real time, doing things that are critically time dependent (perceiving speech, for example). Similarly, humans think the way they do partially because of their size, because of the environment they live in, because of the speed at which they move, live, and think. One of the consistent failings of AI researchers is to vastly underestimate the intricacy and complexity of the kinds of things they are trying to model (of course most of the other cognitive scientists in this century have also underestimated these things). Playing chess is nothing compared with natural language understanding. We take language understanding for granted, because, after all, we all do it. Yet we consider a chess grand master brilliant, because we cannot match his skills. But in fact, becoming a chess grand master is not more difficult than learning to speak and write English. It's easier. We learn language because we start early, spend *lots* and *lots* of time doing it, and it's fun (watch children playing word games sometime). We recognize that it's learn to speak or perish, in a sense. Many fewer people are motivated (at the early age required) to learn to play chess at the GM level. The trouble with the kind of naive (if you'll pardon the expression) reductionism inherent in your position is that it seems to assume that any set of physical interactions that can be expressed mathematically can be scaled up to a full-scale simulation, and that this simulation would be indistinguishable from the original thing. Leaving aside AI for a moment, consider weather simulations. Metereologists have developed computerized simulations of phenomena such as hurricanes. Based on lots of data from past storms, they can predict, with some accuracy, how a developing storm might behave. This is obviously an extremely useful capability. But to claim that a computer simulation of a hurricane is exactly the same as the real thing would probably sound like a very poor joke to someone who has experienced a hurricane first-hand. It seems to me that any intelligent person would say "how could you ever truly simulate a hurricane, and why would you want to?" Well, I have the same reaction to those who say that they want to simulate human intelligence, or even some essential part of it such as natural language understanding. How, and for God's sake, *why*? To study human intelligence, using computers and any other tools available, is a fascinating thing to do. I have spent a number of years doing so. But to say that we are approaching an era when human intelligence will be simulated seems to be just about like saying that from the puff of air generated by the wave of a hand it is only a few short steps to a full-scale realistic simulation of a hurricane. Know what it is you are trying to simulate! -- Tim Smith INTERNET: tsmith@gryphon.CTS.COM UUCP: {hplabs!hp-sdd, sdcsvax, ihnp4, ....}!crash!gryphon!tsmith UUCP: {philabs, trwrb}!cadovax!gryphon!tsmith ------------------------------ Date: 21 Oct 87 05:35:49 GMT From: ucsdhub!jack!man!crash!gryphon!tsmith@sdcsvax.ucsd.edu (Tim Smith) Subject: Re: The Success of AI In article <228@snark.UUCP> eric@snark.UUCP (Eric S. Raymond) writes: +==== | In article <1922@gryphon.CTS.COM>, tsmith@gryphon.CTS.COM (Tim Smith) writes: | > Computers do not process natural language very well, they cannot | > translate between languages with acceptable accuracy, they | > cannot prove significant, original mathematics theorems. | | I am in strong agreement with nearly everything else you say in this article, | especially your emphasis on a need for a new paradigm of mind. But you are, | I think, a little too dismissive of some real accomplishments of AI in at | least one of these difficult areas. | | Doug Lenat's Amateur Mathematician program was a theorem prover equipped with | a bunch of heuristics about what is 'mathematically interesting', essentially | methods for grinding out interesting generalizations and combinations of known | theorems. | [...] | | So at least one of your negative assertions is incorrect. +===== OK, I'll accept your word on this (I'm a linguist, not a mathematician). +===== | I think AI has the same negative-definition problem that "natural | philosophy" did when experimental science got off the ground -- that | once people get a handle on some "AI" problem (like, say, playing | master-level chess or automated proof of theorems) there's a tendency | to say "oh, now we understand that; it's *just* computation, it's not | really AI" and write it out of the field (it would be interesting to | explore the hidden vitalist premises behind such thinking). +===== Well, scientific (and philosophical) fields do progress, and there is a normal tendency to discard the old and no longer interesting. But there is an interesting aspect to what you are saying, I believe. Let me try to develop it a bit, using chess as an example. Chess: I am at a disadvantage here in one sense--I don't play the game very well. In my limited understanding of it, it is a very difficult game to play at a high level. It requires years of study, usually starting at a young age, to become a grand master. It requires peculiar abilities of concentration and nervous resources to play chess at a competetive level. Nevertheless, I don't think of chess as being a particularly intellectual game. It seems much more like tennis to me (and I don't play that either). This is not a put-down! I think of chess as being a sedentary sport--a sport for the mind. Now here's the interesting point. If you were to come to me and say-- "Smith, you have a year to develop an automaton that will play some kind of major sport at a championship level, competing against humans. Money is no object, and you can have access to all the world's experts in AI and robotics, but you must design a robot that plays championship X in a year's time. What is X?" I would say, without a moment's hesistation, "tennis". Why? Of all the sports, tennis is the most bounded. It is played within a very restricted area (unlike golf or even baseball), it is a one-against-one sport (unlike football or soccer), the playing surfaces (aside from Wimbledon) are the truest of all the major sports, and it is indubitably the most boring of all the sports to watch (if not to play). A perfect candidate for automation. Chess? It is tennis for the mind. And so a perfect candidate for initial attempts at AI. But if computers have conquered chess (as they seem about to), does this mean that "real" artificial intelligence is not far behind? No, it just means that chess was over-rated as an intellectual exercise! On a scale of 1 to 10, in terms of intellectual effort involved in playing the game, chess seems to rate at about .002. In terms of skill, concentration ability, depth of understanding of the game, etc. it is difficult. But then, so is multiplying two 37 digit numbers in your head difficult. Unless you're an "idiot savant", or a computer! -- Tim Smith INTERNET: tsmith@gryphon.CTS.COM UUCP: {hplabs!hp-sdd, sdcsvax, ihnp4, ....}!crash!gryphon!tsmith UUCP: {philabs, trwrb}!cadovax!gryphon!tsmith ------------------------------ Date: Wed, 21 Oct 87 09:50:51 PDT From: Tom Dietterich Subject: Lenat's AM program The exact reasons for the success of AM (and for its eventual failure to continue making new discoveries) have not been established. In Lenat's dissertation, he speculated that the source of power was the search heuristics, and that the eventual failure was caused by the inability of the system to generate new heuristics. Then, in a paper by Lenat and Brown, a different reason is given: namely that the representation of concepts was the critical factor. There is a close relationahip between mathematics concepts and lisp, so that mathematical concepts can be represented very concisely as lisp functions. Simple syntactic mutation operations, when applied to these concise functions, yield other interesting mathematical concepts. In new domains, such as those tackled by Eurisko, it was necessary to engineer the concept representation so that the concepts were concisely representable. Finally, in a paper published this year by Lenat and Feigenbaum, yet another explanation of AM's (and Eurisko's) success and failure is given: "The ultimate limitation was not what we expected (cpu time), or hoped for (the need to learn new representations), but rather something at once surprising and daunting: the need to have a massive fraction of consensus reality already in the machine." The problem with all of these explanations is that they have not been subjected to rigorous experimental and analytical tests, so at the present time, we still (more than ten years after AM) do not understand why AM worked! I have my own pet hypothesis, which I am currently subjecting to an experimental test. The hypothesis is this: AM succeeded because its concept-creation operators generated a space that was dense in interesting mathematical concepts. This hypothesis contradicts each of the preceding ones. It claims that heuristics are not important (i.e., a brute force search using the concept-creation operators would be only polynomially--not exponentially--more expensive). It claims that the internal representation of the concepts (as lisp functions) was also unimportant (i.e., any other representation would work as well, because mutation operators are very rarely used by AM). Finally, it claims that additional world knowledge is irrelevant (because it succeeds without such knowledge). There is already some evidence in favor of this hypothesis. At CMU, a student named Weimin Shen has developed a set of operators that can rediscover many of AM's concepts. The operators are applied in brute force fashion and they discover addition, doubling, halving, subtraction, multiplication, squaring, square roots, exponentiation, division, logarithms, and iterated exponentiation. All of these are discovered without manipulating the internal representation of the starting concepts. AM is a "success" of AI in the sense that interesting and novel behavior was exhibited. However, it is a methodological failure of AI, because follow up studies were not conducted to understand causes of the successes and failures of AM. AM is not unique in this regard. Follow-up experimentation and analysis is critical to consolidating our successes and extracting lessons for future research. Let's get to work! Tom Dietterich Department of Computer Science Oregon State University Corvallis, OR 97331 tgd@cs.orst.edu OR tgd%cs.orst.edu@relay.cs.net References: \item Lenat, D. B., (1980). AM: An artificial intelligence approach to discovery in mathematics as heuristic search, In Davis, R., and Lenat, D. B., {\it Knowledge-based systems in Artificial Intelligence}, 1980. \item Lenat, D. B., and Brown, J. S. (1984). Why AM and EURISKO appear to work, {\it Artificial Intelligence}, 23(3) 269--294. \item Lenat, D. B., and Feigenbaum, E. A. (1987). On the thresholds of knowledge. In {\it IJCAI87, The Proceedings of the Tenth International Joint Conference on Artificial Intelligence}, Milan, Los Altos, CA: Morgan-Kaufmann. \item Shen, W. (1987). Functional transformations in AI discovery systems. Technical Report CMU-CS-87-117, Carnegie-Mellon University, Department of Computer Science. ------------------------------ End of AIList Digest ******************** 23-Oct-87 22:13:00-PDT,14732;000000000000 Mail-From: LAWS created at 23-Oct-87 22:06:45 Date: Fri 23 Oct 1987 21:56-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #243 - Lisp Text, Prolog, Design, Cash Flow, Neural Nets To: AIList@SRI.COM AIList Digest Saturday, 24 Oct 1987 Volume 5 : Issue 243 Today's Topics: Query - Connection Machine Architecture & Neural Info Process Conference & Multiexpert/Multiagent Researches & Prolog for Course, Education - Common Lisp Textbooks & Prolog, Application - AI and Design Automation/Design Assistance & Cash Flow, Neuromorphic Systems - Byte Sources & Cybernetics ---------------------------------------------------------------------- Date: 21 Oct 87 19:16:18 GMT From: Mark Attisha Subject: Info on Connection Machine Wanted We are wondering if anyone can provide us with information specific to the Connection Machine. We have found that Hillis' book to be lacking in such areas as hardware description and network communications. In particular, we are interested in obtaining a description of processing element to router communication, host to processing element communication, chip control unit to processing element communication, the role of shared memory, a description of the buses between chips and between processing elements, and so forth. Thanks in advance. Please send information to the: Mark Attisha Department of Computing & Information Science Queen's University Kingston, Ontario K7L 3N6 Canada e-mail attisha@qucis.bitnet ------------------------------ Date: 23 Oct 87 01:49:55 GMT From: deneb.ucdavis.edu!g451252772ea@ucdavis.ucdavis.edu (0040;0000001585;0;327;142;) Subject: Neural Info Process conf. at Denver 11/8-12 Having just been informed I have funds to attend this, it would be gratifying to learn if it's still open (moving to Denver changed the crowd capacity to infinity, yes?) I'm also interested if anyone has ideas on lodgings less expensive than the Sheraton ... or travel inexpensively from near the Bay Area to Denver (Davis is closest to Sacramento physically, but to SF otherwise...) --thanks! Ron Goldthwaite / UC Davis, Psychology and Animal Behavior 'Economics is a branch of ethics, pretending to be a science; ethology is a science, pretending relevance to ethics.' (apologies if the signature appears 2x) ------------------------------ Date: 23 Oct 87 18:14:28 GMT From: leey@russell.STANFORD.EDU (Chin Lee) Reply-to: leey@russell.stanford.edu (Yi-Chin Lee) Subject: Need pointers to multi expert -- multi agent researches Is there anyone out there on the net can provide me with bibliographical pointers to researches related to multi expert -- multi agent planning and knowledge representation? Thanks. ------------------------------ Date: 22 Oct 87 23:50:08 GMT From: ucsdhub!hp-sdd!ncr-sd!ncrcae!hubcap!steve@sdcsvax.ucsd.edu ("Steve" Stevenson) Subject: Suggestions for Course I have to teach an AI course for folks with little or no background. I'd like to use prolog, but want to have them learn it as much on their own as possible. Any suggestions for texts? At this time, I think I would like to concentrate on theorem proving with perhaps some non-traditional stuff (fuzzy?) included. Any suggestions here? -- Steve (really "D. E.") Stevenson steve@hubcap.clemson.edu Department of Computer Science, (803)656-5880.mabell Clemson University, Clemson, SC 29634-1906 ------------------------------ Date: Thu, 22 Oct 87 10:50:40 EDT From: Chris Riesbeck Subject: AI Programming Book Date: 19 Oct 87 10:54:01 edt From: Walter Hamscher Subject: Introductory books on Lisp Charniak, Riesbeck, McDermott "Artificial Intelligence Programming" Lawrence Erlbaum (1980). What every AI programmer should know, though unfortunately the lisp dialect is getting a bit dated. The Second Edition is now available, in Common Lisp, substantially revised, with old bugs and typos replaced by sparkling new ones. Artificial Intelligence Programming, Second Edition (1987) Charniak, Riesbeck, McDermott, and Meehan Lawrence Erlbaum Assoc, Inc. 365 Broadway, Hillsdale, NJ 07642 ------------------------------ Date: Thu, 22 Oct 87 08:21:00 PDT From: Marie Bienkowski Subject: Common Lisp Textbooks Date: Fri, 16 Oct 87 11:03:46 PDT From: glasgow@marlin.nosc.mil (Michael G. Glasgow) I am new to AIList and AI programming and want to learn Lisp. I have been looking through Steele's book, Common Lisp", and have discovered that this is more of a reference manual than a beginners guide. What I am wondering is if anyone can give me the names of some good introductory Lisp books to get me started. As several people mentioned in response to this query, there are several good texts on Common Lisp. What surprises me is that no one mentioned Deborah Tatar's book. First, let me say that I relied on Winston and then Wilensky when teaching LISP, both are good books. (with Wilensky's being better). But when I tried to learn Common Lisp from Steele, then found it was impossible, I discovered Tatar's excellent book. It's published by Digital Press, and is entitled "A Programmer's Guide to Common Lisp." While I have not used her book for teaching, I think the examples are good enough to warrant its use. And it is the perfect companion to Steele (in fact, Steele, wrote the foreward for it). It may be more difficult to get than, say, Wilensky's, but I think it is worth it. (If you go for nice-looking covers, on the other hand, get Wilensky's. It's great-looking!) Marie Bienkowski bienk@istc.sri.com ------------------------------ Date: 23 Oct 87 16:25:20 GMT From: oltz@tcgould.tn.cornell.edu (Michael Oltz) Reply-to: oltz@tcgould.tn.cornell.edu (Michael Oltz) Subject: Re: Introductory books on Lisp In article <8710191454.AA26556@ht.ai.mit.edu> hamscher@HT.AI.MIT.EDU (Walter Hamscher) writes: > Charniak, Riesbeck, McDermott "Artificial Intelligence > Programming" Lawrence Erlbaum (1980). What every AI programmer > should know, though unfortunately the lisp dialect is getting a > bit dated. At a talk McDermott gave at Cornell in September, it was announced that the 2nd edition of this book would be coming out soon. -- Mike Oltz oltz@tcgould.tn.cornell.UUCP (607)255-8312 Cornell Computer Services 215 Computing and Communications Center Ithaca NY 14853 ------------------------------ Date: 23 Oct 87 14:15:52 GMT From: ucsdhub!hp-sdd!ncr-sd!ncrlnk!ncrcam!morley@sdcsvax.ucsd.edu (/usr/acct/morley) Subject: Re: Suggestions for Course In article <587@hubcap.UUCP>, steve@hubcap.UUCP ("Steve" Stevenson) writes: > I have to teach an AI course for folks with little or no > background. I'd like to use prolog, but want to have them > learn it as much on their own as possible. Any suggestions > for texts? How about Turbo Prolog? Some will argue that it is not "true" Prolog, but it is very close to the real thing. The manual is in tutorial form, and is easy to learn and use. Also, Borland International offers a discount to students. The price is very reasonable. > -- > Steve (really "D. E.") Stevenson steve@hubcap.clemson.edu > Department of Computer Science, (803)656-5880.mabell > Clemson University, Clemson, SC 29634-1906 ------------------------------ Date: 23 Oct 87 19:13:51 GMT From: bbn!gatech!hubcap!grimlok@husc6.harvard.edu (Mike Percy) Subject: Re: Suggestions for Course in article <321@ncrcam.Cambridge.NCR.COM>, morley@ncrcam.Cambridge.NCR.COM (/usr/acct/morley) says: > Xref: hubcap comp.lang.prolog:345 comp.ai:820 > > In article <587@hubcap.UUCP>, steve@hubcap.UUCP ("Steve" Stevenson) writes: >> I have to teach an AI course for folks with little or no >> background. I'd like to use prolog, but want to have them >> learn it as much on their own as possible. Any suggestions >> for texts? > > How about Turbo Prolog? Some will argue that it is not "true" Prolog, but > it is very close to the real thing. The manual is in tutorial form, and is > easy to learn and use. Also, Borland International offers a discount to > students. The price is very reasonable. > True about TProlog, it is almost Prolog, but not quite. In fact, at some places it is downright divergent and unusable. But for the environment Dr. Stevenson is in, nearly every one of his students has used at least Turbo Pascal and possibly TurboC. They are familiar with the Borland systems, and can concentrate on their programs rather than than their compiler and how to use it. Also, the speed of testing is nice, the debugging trace is helpful, and Clemson has plenty of PCs. In these days, when the VAX machines are quickly becoming overloaded, any PC implementation will be a plus. So Dr. Stevenson, here is my vote for TProlog, with the proviso that you declare to the poor students that TProlog is a mere shadow of the true power of the language. Mike Percy Clemson University ------------------------------ Date: 22 Oct 87 17:14:00 EDT From: "ETD1::WILSONJ" Reply-to: "ETD1::WILSONJ" Subject: AI & Design Automation, Design Assistance I'm beginning a study on Applications of AI in Design Automation and Design Assistance. My interests range from IC CAD (my area of expertise) to the design of mechanical, aerodynamic, and propulsion systems; and beyond. I need to explore today's design issues, where does AI fit in, what are the most critical design needs. I'd greatly appreciate brief replys on who's doing what in AI & Design, and issues that you believe should be persued, i.e. the most promising of advances in design, where is research lacking, etc. I presently work in an AI prototyping facility whose function is to rapidly transition state of the art AI technology into Air Force weapon systems; and serve as a showcase for state of the art AI applications research and the latest AI hardware and software innovations. My research will help direct training and technical efforts at a newly established AI Applications Center at the Miami Valley Research Institute (MVRI) in Dayton, OH. Aeronautical Systems Division/Air Force Systems Command at Wright Patterson AFB, OH awarded a $10 million contract to MVRI last week. MVRI is a consortium of the University of Dayton, Wright State University, Central State University, and Sinclair Community College. Teknowledge Federal Systems Division and the Ohio State University Laboratory for Artificial Intelligence Research (LAIR) will serve as subcontractors. Thank You, :-) James B. Wilson ARPANET: wilsonj%etd1.decnet@ US Mail: AFWAL/AAI Bldg 22, Area B WPAFB, OH 45433 Phone: (513) 255-1491 ------------------------------ Date: Thu, 22 Oct 87 11:01:28 EDT From: Brady@UDEL.EDU Subject: Lopez query To respond to Javier Lopez' query: 1. a couple of years ago AI Magazine had an article specifically on this subject. The author likened the flow of cash through a company to the flow of water into, through, and out of a system of pipes and valves. The discussion reminded me of systems that model electrical circuitry. 2. Most intermediate accounting and finance texts treat the cash flow concept well. Usually, the most vexatious problem in predicting cash inflows is estimating revenues, so a good market analysis text may also help. 3. IEEE Expert recently devoted an issue to financial applications. There may be something there about your topic. ------------------------------ Date: Thu 22 Oct 87 13:58:44-PDT From: Matt Heffron Subject: Re: neuro files For general Info: The sources to programs from Byte magazine are available from the BYTEnet bulletin board system: (617)861-9764 (set modem at: 8-1-N or 7-1-E; 300 or 1200 baud). This system supports several PC "ftp" protocols, including xmodem (and "standard" variations...) -Matt Heffron BEC.HEFFRON@ECLA.USC.EDU ------------------------------ Date: 21 Oct 87 13:59:56 GMT From: trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU (Stephen Smoliar) Subject: Re: Neural Networks - Pointers to good In article <8300006@osiris.cso.uiuc.edu> goldfain@osiris.cso.uiuc.edu writes: > > On the other hand, whatever became of the term "cybernetics" that Norbert >Weiner coined long ago? I thought its definition was quite suitable for >denoting this research. I do not profess to be an expert in either the history of cybernetics or the usage of the term; but, with that qualification, let me try to address this question. As I recall, Weiner's original concern was with the design of analog devices which, by virtue of feedback circuits, were capable of control of other devices and adaptive behavior (which may be regarded as self-control). Through my encounters with the literature as an AI researcher, I have observed that the term "cybernetics" appears with greater frequency in Europe (particularly the Soviet Union and the United Kingdom) than it does in the United States. There is definitely a tendency to recognize that Weiner's original principles could be generalized from analog to digital hardware. However, I have the distinct impression that cybernetics grew from the belief that behavioral knowledge was something which would ultimately be encoded in the feedback loops, rather than in an explicit device concerned with memory or the storage of a knowledge base. I would appreciate any reactions to these comments simply to get the historical record straight. ------------------------------ End of AIList Digest ******************** 23-Oct-87 22:19:41-PDT,24119;000000000000 Mail-From: LAWS created at 23-Oct-87 22:17:50 Date: Fri 23 Oct 1987 22:15-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #244 - Financing, Neuromorphic Terminology, Flawed Minds To: AIList@SRI.COM AIList Digest Saturday, 24 Oct 1987 Volume 5 : Issue 244 Today's Topics: Business - Expert Systems Company Financing, Neuromorphic Systems - Terminology & Textbook, Philosophy - Flawed Human Minds ---------------------------------------------------------------------- Date: 18 Oct 87 17:52:45 GMT From: trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU (Stephen Smoliar) Subject: Re: Expert Systems Company Financing... In the early eighteenth century a man of intense religious fervour named Johann Ernst Elias Bessler claimed that God had revealed to him the secret of the perpetual motion machine. He would tour villages in the costume of a magician and offer demonstrations of his devices. Ultimately, he attracted the attention of Count Karl von Hessen-Cassel, who undertook to serve as a sponsor. At Hessen-Cassel's expense, Bessler built one of these machines based on a wheel which was twelve feet in diameter. Hessen-Cassel then invited many of the leading scientific minds of his time to evaluate the project. In the course of this evaluation, the machine apparently ran without stopping for 54 days. Ultimately, Bessler was exposed as a fraud; and several scientific reputations were destroyed as a consequence. While the historical record of this affair is fragmented, there are several rather interesting points which I would claim are at least remotely related to the current discussion about similar sponsorship of artificial intelligence. 1. The evaluating scientists were not allowed to inspect the inner workings of Bessler's machine. Bessler claimed they would be blinded by the divine revelation (or words to that effect). Hessen-Cassel apparently did see the inner workings and was not blinded. Nevertheless, the evaluating committee agreed to accept this constraint. 2. For all the time that Hessen-Cassel possessed this machine, he never tried to do anything practical with it. Bessler's previous demonstrations with smaller-scale machines always climaxed with the machine being used to lift some impressive weight. While Hessen-Cassel was in possession of a potentially significant labor-saving device, he seemed content to keep it locked in a room of his castle. 3. Bessler was never exposed on the grounds of any scientific argument. Willem Jakob Gravesande published a "proof" of why the machine worked, and the flaw in this proof was subsequently published by Jacque de Crousaz. However, Bessler was undone when a servant girl confessed that she was powering the machine from an adjoining room. This was later discovered to be a false testimony, but Bessler was distraught by the affair. Before anyone had a chance to inspect its interior, he destroyed the machine. I do not intend to imply that artificial intelligence is like perpetual motion, at least to the extent that it is a theoretical impossibility. However, I am struck by certain behavioral parallels between past and present. My personal opinion is that Bessler was probably an extremely skilled "hacker" (in mechanics) for his time, with his personal confidence reinforced by his religious convictions. He probably pulled off a pretty good piece of work even if his mind was entirely "in the bits" (so to speak) and largely ignorant of prevailing theory. What is pathetic, however, is that those who were asked to evaluate him were willing to play the game by his own rules. Indeed, there is some indication that their opinions may have been slanted by the promise of sharing in the monetary gain which Bessler's invention might yield. Also, there is this depressing observation that the evaluation never involved putting the machine to work; they were content to just let it run on in a locked chamber. Current "success stories" about artificial intelligence are not quite as contrived as that of Bessler's machine running in a locked room for 54 days; but they come closer than I would feel is comfortable. To a great extent, the "field testing" of "applied" expert systems often takes place in rather constrained circumstances. A less polite way of putting this might be to say that the definition of "success" is in danger of being modified POST HOC to accommodate the capabilities of the system being evaluated. Thus, I feel that all reports of such stories should be viewed with appropriate scientific scepticism. On the other hand, there is a positive side of this historical retrospective. Had Hessen-Cassel actually put Bessler's machine to work, it might have been of considerable benefit to him . . . even if it did not run forever. In other words, a machine capable of dissipating its energy slowly enough to run for a very long time, while not being a true perpetual motion machine, would still be a useful tool. By concentrating on a theoretical goal, rather than a practical one, Hessen-Cassel lost an opportunity to exploit a potentially valuable resource. Similarly, sponsorship of artificial intelligence should probably pay more heed to advancement along specific pragmatic fronts and less to whether or not machines which exhibit that behavior deserve to be called "intelligent." If we recognize what we have for what it is, we may get more out of it than we might think. ACKNOWLEDGEMENT: I would like to thank Jim Engelhardt for the extensive research he has performed regarding the story of Bessler. He is in the process of incorporating his research into a play which he is calling THE PERPETUAL MOTION MAN. His research has been quite thorough, and his insights are noteworthy. ------------------------------ Date: Mon, 19 Oct 87 09:29 EST From: "William E. Hamilton, Jr." Subject: RE: AIList V5 #239 - Neuromorphic Terminology, AI Successes, Of course the human mind is flawed. The proof is quite straightforward. 1. The Bible asserts that the mind of man is wicked (or evil, depending on which translation you use) 2. Now, assuming you believe the Bible is true and that an evil mind is a flawed mind (if you don't agree that an evil mind is flawed, you can still find a collection of assertions about human behavior in the Bible that, taken together, would indicate that the minds responsible for such behavior are flawed), the assertion is proven. 3. But suppose you do not regard the Bible as true. Then the Bible is flawed. However, the Bible has been on the world's best-seller list for many years, and those who buy a flawed book must have flawed minds. Therefore, there are millions of flawed minds out there. Bill Hamilton GM Research Labs ------------------------------ Date: 19 Oct 87 07:27:00 GMT From: uxc.cso.uiuc.edu!osiris.cso.uiuc.edu!goldfain@a.cs.uiuc.edu Subject: Re: Neural Networks - Pointers to good I agree with the respondent whose user-name was listed as "smoliar" that this haggling about earliest references is "silly". In fact, I don't understand the need for any territorial fight over terminology here. Do physiologists actually use the two-word term "neural network" in their literature? "Neuron", and "neural tissue", surely, but do they actually use "neural network" ? If not, then there is no ambiguity. Sure there is some danger of confusion, but no more than I think is usual in cases of "learned borrowing". The term "neural network" as used by "connectionist/AI" researchers caught on precisely because this model of computation is based on the gross behavior of real, mammalian-brain neurons. It can be viewed in some ways as a study of the human brain itself. Thus it is no greater an abuse of terminology than, for example, "pipeline computers". On the other hand, whatever became of the term "cybernetics" that Norbert Weiner coined long ago? I thought its definition was quite suitable for denoting this research. I doubt that "connectionist" is much help, in view of the fact that the "connection machine" is more a project in pure parallelism than intended as a neural model. If I am wrong about any of this, please enlighten me. ------------------------------ Date: 21 Oct 87 15:49:02 GMT From: ssc-vax!dickey@beaver.cs.washington.edu (Frederick J Dickey) Subject: Re: Neural Networks - Pointers to good texts? From postnews Wed Oct 21 07:49:49 1987 > >interpretation, let me make the following observation. In 1943, McCulloch > >and Pitts published a paper entitled "A logical calculus of the ideas > >immanent in neural nets". Minsky and Papert (Perceptrons) state that this > > Well . . . this is all rather silly. The PUBLISHED title of the classic > paper by McCullogh and Pitts is "A Logigal Calculus of the Ideas Immanent > in Nervous Activity." They NEVER use "neural net" as a technical term Well, this is very interesting. When I read Calvin's original posting I was struck by the claim that neural nets had been studied for 25 years. This surely seemed too small a figure to me. To check this out without initiating a major research project, I grabbed a copy of Minsky and Papert's "Perceptrons" which happened to be on my desk at the time and opened to the bibliography. M&P give the title of the McCullough and Pitts paper as "A logical calculus of the ideas immanent in neural nets". I'm looking at it right now and that's what it says. Apparently, the citation is wrong. Well, I stand corrected. I might comment by the way that regardless of the merits of Calvin's claim that artificial neural nets ought to be named something else, I think the effort is doomed to failure. The reason being that we seem to have become an excessively marketing-oriented society. Labels are attached to things to stimulate desired responses in "consumers," not to clarify crtical distinctions. The practical problem one faces in much of the industrial world is attempting to gain support for one's "research." To do this, one presents one's proposed program to a manager, i.e., one markets it. The noise level in this process is so high that nothing less than hype makes it through. My experience with managers leads me to believe that they may have heard of neural nets. If I tried to start a "neuroid" project, they would say "Isn't that the same thing as a neural net?" I can guarantee you that they aren't interested in the distinctions between artifical and biological nets. How can an aerospace company make a profit from biological nets? In other words, to start a artifical neural net project, I have to call it a neural net, show how it applies to some product, how it adds value to the product (neural nets will make the product more powerful than a locomotive, faster than a speeding bullet, and able to leap over tall buildings at a single bound), and how all this can be done by next year at a half man-year level of effort. If I lived in an ivory tower (a not unpleasant domicile), I'd say that Calvin is right on. Out here in the cinder block towers, he's out to lunch. To summarize, I'm sympathic to his viewpoint, but my sympathy isn't going to make much difference. ------------------------------ Date: 23 Oct 87 06:50:34 GMT From: well!wcalvin@LLL-LCC.ARPA (William Calvin) Subject: Why "neural nets" is a bad name I admit that "nerve nets" and the variant "neural networks" are catchy titles; we neurobiologists have used the terms quite a lot, though mostly informally as in the annual meeting called the "Western Nerve Net". Each real neural network tends to become its own subject name, as in "stomatogastric" and "retina", with papers on properties that transcend particular anatomies incorporated into sessions called "theoretical neurobiology" or some such (I'm on the editorial board of the JOURNAL OF THEORETICAL NEUROBIOLOGY, often concerned with networks). A quarter-century ago was the era of the Perceptron, the first of the network learning models. Various people were simulating network properties using neuron-like "cells" and known anatomy; when I was a physics undergrad in 1959, I did an honors thesis on simulating the mammalian retina (using anatomy based only on light-microscopy, using physiology of neurons borrowed from cat spinal motorneurons, using sensory principles borrowed from horseshoe crab! A far cry from the CRAY-1 simulations these days using modern retinal neurobiology). And if you think that your simulations run slow: I did overnight runs on an IBM 650, which had to fetch each instruction from a rotating drum because it lacked core memory. Now this was also the era when journalists called any digital computer a "brain" -- and I've pointed out that calling any pseudo-neural network a "neural network" is just as flaky as that 60s journalistic hype. Now brain researchers were not seriously inconvenienced by the journalistic hype -- but I think that blurring the lines is a bad idea now. Why? Real neural networks will soon be a small part of a burgeoning field which will have real applications, even consumer products. To identify those with real brain research may seem innocuous to you now because of the frequent overlap at present between pseudo-neural networks and simulations of real neural circuitry. But these distributed networks of pseudo-neurons are going to quickly develop a life of their own with an even more tenuous connection to neuroscience. They need their own name, because borrowing is getting a bad name. Let me briefly digress. We are already seeing a lot of hype based on a truly nonexistent connection to real neuroscience, such as those idiot "Half Brain or Whole Brain" ads in the Wall Street Journal and New York Times, where "John-David, Ph.D." describes himself as one of the "world's most recognized neuroscientists" recently "recognized as a Fellow by the International Institute of Neuroscience" (Nope, I've never heard of it either, and I was a founding member of the Society for Neuroscience back in 1970). See James Gorman's treatment in DISCOVER 11/87 p38. Is this just feel-good floatation- tank pseudo-psychology dressed up to look like hard science, another scheme to part half-brained fools from their money? Scientists are going to start to get touchy about consumer products borrowing an inferred plug from real science, just as the FDA has gotten touchy about white coats in Carter's Little Liver Pills advertisements attempting to convey medical approval. And you can bet that, if pseudo-neural nets become as successful as I think they will, some advertising genius will try to pass off a nonfunctional product as a neural network "resonating with your brain", try to get some of that aura of real science and technology to rub off on the sham. Do you really want your field trapped in the middle of an FDA/FTC battle with the sham exploiters because it initiated the borrowing? Borrowing a name for a technology from a basic science is not traditional: civil engineers do not call themselves "physicists". We neurobiologists are always having to distinguish the theoretical possibilities, such as retrograde transport setting synaptic strengths, from reality. Those theoretical possibilities may, of course, be excellent shortcuts that Darwinian evolution never discovered. And so we'll see distinctions having to be drawn: "backpropagation works in pseudo-neural nets, but hasn't been seen so far in real neural nets." If you call the technology by the same name as the basic science, you start confusing students, journalists, and even experienced scientists trying to break into the field -- just try reading that last quote with "pseudo" and "real" left out. William H. Calvin University of Washington NJ-15 Seattle WA 98195 wcalvin@well.uucp wcalvin@uwalocke.bitnet 206/328-1192 206/543-1648 ------------------------------ Date: 18 Oct 87 23:37:35 GMT From: ctnews!pyramid!prls!philabs!gcm!dc@unix.sri.com (Dave Caswell) Subject: Re: Flawed human minds >> Factually, we know the mind is flawed because we observe that it does >> not do what we expect of it. > Factually, the mind knows the mind is flawed because the mind observes the mind not doing what the mind expects the mind to do. ------------------------------ Date: 20 Oct 87 17:43:26 GMT From: ucsdhub!hp-sdd!ncr-sd!ncrlnk!ncrday!seradg!bryan@sdcsvax.ucsd.ed u (Bryan Klopfenstein) Subject: Re: Flawed human minds In article <359@white.gcm> dc@white.UUCP (Dave Caswell) writes: >>> Factually, we know the mind is flawed because we observe that it does >>> not do what we expect of it. >> >Factually, the mind knows the mind is flawed because the mind observes the >mind not doing what the mind expects the mind to do. So, is the mind flawed because it expects the wrong thing, or is the mind flawed because it observes incorrectly, or is the mind flawed because it does not live up to its expectations? Or is this a ridiculous question and a flawed mind does not have the capability to evaluate itself, thus making it unable to determine whether or not is really is flawed? -- Bryan Klopfenstein CSNET bryan@seradg.Dayton.NCR.COM NCR Corporation ARPA bryan%seradg.Dayton.NCR.COM@relay.cs.net VOICE (513) 865-8080 -- Standard Disclaimer Applies -- ------------------------------ Date: 20 Oct 87 15:43:54 GMT From: ihnp4!homxb!genesis!odyssey!gls@ucbvax.Berkeley.EDU (g.l.sicherman) Subject: Re: The Job Hunt > Do we need a definition of anger? Anger, as I understand it, is an > emotion that catalyzes physical actions but interferes with reason. > I agree that Mr. X may rationalize his action, but I don't believe > it was his best choice. ... > > ... I thought what we all needed was a little humility. If > Col. G. L. Sicherman thinks either that he is perfect, or that I am > perfect, I disagree. Tentatively. If you go telling people what you think they all need, we may decide that you're not very humble! Arguing over whether people are "perfect" or "flawed" is like arguing whether Eugene the Jeep is a rodent or a marsupial. Perfect for *what?* And I agree that we need a definition of anger. "Catalyzes physical actions?" The anger *produces* the actions. If you had no emotions, you would never act. > ... I believe that at least some emotional responses are > maladaptive and would not exist in a perfect intelligence, while he > apparently believes the human mind is perfect and cannot be improved > upon. Again, perfect for what? It sounds as if you regard emotions as a part of intelligence. We don't agree on the basics yet. "This rock, for instance, has an I.Q. of zero. Ouch!" "What's the matter, Professor?" "It bit me!" -- Col. G. L. Sicherman ...!ihnp4!odyssey!gls ------------------------------ Date: Tue, 20 Oct 87 16:21:42 PDT From: larry@VLSI.JPL.NASA.GOV Subject: Flawed/Flawless FLAWED/FLAWLESS: I can argue on both sides. FLAWLESS: Quality can only be judged against some standard. Every person has (perhaps only slightly) a different value system, as may even the same person at different times. So what is a flaw to one may be a "feature" to another. (Examples: a follower of Kali may consider torture, death, and corruption holy; a worshipper of the Earth Mother may consider monogamy a sin and infertility a crime.) The only objective standard is survival of the largest number of one's species for the longest time, and even this "standard" is hopelessly flawed(!) by human subjectivity. FLAWED: Nevertheless, humans DO have standards, and not only for made objects like automobiles, that are essential to our survival and happiness. We want lovers who have compatible timing (social, sexual), sensitivity (at least enough not to hog the blankets or too obviously eye the competition), enough intelligence (so we can laugh at the same jokes) but not too much (winning an occasional argument is necessary to our self-esteem), etc. Notice that TOO MUCH intelligence may be considered as bad a flaw as too little. And more FLAWLESS: From an evolutionary standpoint what is a "virtue" in one mileau may become deadly when the environment changes. Performing some mental activity reliably may be of little use when chaos sweeps through our lifeways. THEN divergent thinking--or even simple error--may be more likely to solve problems. A perfect memory (popularly thought to accompany great intelligence) can be a liability, holding one rigidly to standards or knowledge no longer valid. It is also the enemy of abstract/general thought, which depends on forgetting (or ignoring) inessentials. (Indeed, differential forgetting may be one of those great ignored areas of fruitful research.) AI: What does all this have to do with artificial intelligence? Maybe nothing, but I'll invent something. Say ... the relationship of emotions to intelligence. First, sensations of pain and pleasure drive thought, in the sense that they establish values for thinking and striving to achieve or avoid some event or condition. Sensation triggers emotions and in turn triggers sensations which may act as second-level motivators. They also may trigger subsystems for readiness (or inhibition) of action. (Example: hunger depletes blood sugar, triggering anger when a certain level of stress is reached, which releases adrenalin which energizes the body. Anger also may cause partly or fully random action, which is statistically better than apathy for killing or harvesting something to eat.) At least that's the more traditional outlook on emotions--though that outlook may have changed in ten years or so since I did much reading in psychology. Even if true, however, the above outlook doesn't establish a necessary link of emotion with artificial thought; humans can supply the goals and values for a Mars Rover, an activate command can trigger emergency energy reserves. Some other more intimate association of emotion with thought is needed. Perhaps emotions affect thought in beneficial ways, say improving the thinking mechanism itself. (The notion that it impedes correct thought is too conventional and (worse) obvious to be interesting.) Or maybe emotion IS thought in some way. It is, after all, only a conclusion based on incomplete brain evidence that thought is electrical in nature. Suppose the electrical action of the brain is secondary, that the biochemical action of the brain is the primary mechanism of thought. This might square with the observation that decision often happens in the subconscious, very rapidly, and integrates several (or even dozens) of conflicting motives into a vector sum. In other words, an analog computer may be a better model for human thought than a digital one. (In the nature of things that answer is likely too simple. Most likely, I'd guess, the brain is a hybrid of the two.) Larry @ jpl-vlsi ------------------------------ End of AIList Digest ******************** 23-Oct-87 22:32:01-PDT,21245;000000000000 Mail-From: LAWS created at 23-Oct-87 22:28:37 Date: Fri 23 Oct 1987 22:23-PDT From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #245 - AM, Success of AI, Dreyfus's Philosophy To: AIList@SRI.COM AIList Digest Saturday, 24 Oct 1987 Volume 5 : Issue 245 Today's Topics: Comments - Lenat's AM & The Success of AI & Dreyfus's Philosophy ---------------------------------------------------------------------- Date: 20 Oct 87 16:14:48 GMT From: trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU (Stephen Smoliar) Subject: Re: The Success of AI In article <9320@ut-sally.UUCP> brian@ut-sally.UUCP (Brian H. Powell) writes: > If you read the summaries of AM, you think it's powerful. Once you read >the entire dissertation, you realize it's not quite as great a program as you >had thought, but you still think it's good research. > Actually, Lenat and John Seely Brown did something rather like this when they wrote the paper "Why AM and Eurisko Appear to Work" for AAAI-83. ------------------------------ Date: Thu 22 Oct 87 08:35:35-PDT From: Douglas Edwards Subject: Reworking Lenat The claim that Lenat's work has not been retested should not be allowed to pass without being questioned. Not only should Weimin Shen's work, already cited by Tom Dietterich, be taken into account, but there is apparently another attempt to work with the same approach going on at MIT. *Artificial Intelligence Abstracts* cites the MIT AI Memo AIM-898, "Discovery Systems" by K. W. Haase Jr. (*AI Abstracts*, volume 1 number 1, January 1987). I have not yet read (or even obtained) this memo, but the abstract suggests that Haase has not only reimplemented Lenat's work but also tried to discover a principled explanation for why it works, and that Haase's explanation for AM's success would be quite different from Dietterich's and Shen's. I look forward to learning more about Haase's work. I don't know if Haase reads AILIST; if he does, it would be interesting to hear his own comments on the AM controversy. --- Douglas D. Edwards (edwards@ai.sri.com) ------------------------------ Date: 20 Oct 87 14:48:35 GMT From: bpa!cbmvax!snark!eric@burdvax.prc.unisys.com (Eric S. Raymond) Subject: Re: The Success of AI In article <9320@ut-sally.UUCP>, brian@ut-sally.UUCP (Brian H. Powell) writes: > I feel compelled to challenge this, but not necessarily the rest of your > article. > AM wasn't a theorem prover. From the July, 1976 dissertation: Thanks for the correction, which I also received by email from another comp.ai regular. I never saw Lenat's dissertation, just an expository paper in one of journals. I guess maybe the reason I thought the sucker had a theorem prover attached was that I was working on LISP support for a theorem prover at the time, and my associative memory got a collision in its hash tables :-). Nevertheless, I think my more general observations about AI's definitional problem remain valid. Compilers are a 'success' of AI. So are heuristic-based search-and-backtrack algorithms. So is the visual analysis preprocessing used in seeing pick-and-place robots. So (most recently) are 'expert systems'. In *each case*, these problem areas were defined out of the AI field as soon as they spawned halfway-usable technologies and acquired their own research communities. I think the same thing is about to happen to neural nets, BTW... -- Eric S. Raymond UUCP: {{seismo,ihnp4,rutgers}!cbmvax,sdcrdcf!burdvax,vu-vlsi}!snark!eric Post: 22 South Warren Avenue, Malvern, PA 19355 Phone: (215)-296-5718 ------------------------------ Date: 21 Oct 87 23:33:18 GMT From: psuvax1!vu-vlsi!swatsun!scott@rutgers.edu (Jay Scott) Subject: Re: The Success of AI > Nevertheless, I think my more general observations about AI's definitional > problem remain valid. Compilers are a 'success' of AI. So are heuristic-based > search-and-backtrack algorithms. So is the visual analysis preprocessing used > in seeing pick-and-place robots. So (most recently) are 'expert systems'. > In *each case*, these problem areas were defined out of the AI field as soon > as they spawned halfway-usable technologies and acquired their own research > communities. I agree. And I want to understand better why it's so. Here's one speculation: People see intelligence as mysterious, intrinsically non-understandable. So anything understood can't be part of intelligence, and can't be part of AI. I assume this was what Eric had in mind in a previous article, when he mentioned "hidden vitalist premises". Of course some people believe explicitly that intelligence is mystical, and say so. But even AI people may implicitly feel that, oh, this algorithm isn't clever enough, real intelligence has to be cleverer than that. And so it goes. Any other good ideas? > Eric S. Raymond > UUCP: {{seismo,ihnp4,rutgers}!cbmvax,sdcrdcf!burdvax,vu-vlsi}!snark!eric > Post: 22 South Warren Avenue, Malvern, PA 19355 Phone: (215)-296-5718 Jay Scott ...bpa!swatsun!scott ------------------------------ Date: 20 Oct 87 16:04:07 GMT From: trwrb!aero!venera.isi.edu!smoliar@ucbvax.Berkeley.EDU (Stephen Smoliar) Subject: Re: The Success of AI (Analysis of AI lack of progress). Those who would like a taste of the Dreyfus style before embarking upon one of his books in its entirely would do well to consult the Summer 1986 issue of IEEE EXPERT. The article "Why Expert Systems Do Not Exhibit Expertise," by Hubert and Stuart Dreyfus, is an excerpt from MIND OVER MACHINE: THE POWER OF HUMAN INTUITION AND EXPERTISE IN THE ERA OF THE COMPUTER. While there is definitely merit to deflating exaggerated claims about expert systems which have been made in the name of salesmanship, Hubert Dreyfus approaches this issue as a philosopher. Consequently, the technical baggage he carries is often not particularly timely and often inadequate. Were he to wage his campaign on the battelground of the philosophy of mind, he might come away with some notable victories; but by descending to the level of technology, he often falls into traps of misconception. Here is a sample passage: Humans often think by forming images and comparing them holistically. This process is quite different from the logical, step-by-step operations that logic machines perform. There are several things wrong here. First of all, a holistic theory of memory or reasoning remains a HYPOTHESIS. Claiming it as an observation is a gross misrepresentation of the surrent state of cognitive science. Second, the term "logic machine" has been introduced to capture a particular machine architecture which lacks what Dreyfus wants it to lack. He does not admit of the possibility of an alternative architecture for the mechanization of thought which could model the holistic hypothesis. Fortunately, more productive cognitive scientists HAVE pursued this line of reasoning. In any event, the text continues in an attempt to elaborate upon this point: For instance, human beings use images to predict how certain events will turn out. This is, again, hypothesis. It rests on a weaker hypothsis which is never cited: that human beings use MODELS to predict how certain events will turn out. This is the whole "mental models" approach to cognition, for which there is both subtantial literature and experiments in mechanical implementation. The text continues: Programming a computer to analyze a scene has turned out to be very difficult. Such programs require a great deal of computation, and they work only in special cases with objects whose characteristics the computer has been programmed to recognize in advance. Nevertheless, such programs may work better than people in those special cases and can be used in factories. That is why industrial robotics has become as effective as it has. I regard this as an instance of the situation I raised regarding perpetual motion machines in an earlier note. I raised the point that had Bessler's machine actually been put to work and found to run for significantly long periods of time without energy input, it would have been an impressive contribution even if its energy dissapated very slowly, rather than not at all. Similarly, we would do better to study special cases of scene analysis which are successes rather than belabor the obstacles to a more general approach to the task. It gets better: But that is just the beginning of the problem. Computers currently can make inferences only from lists of facts. It's as if to read a newspaper you had to spell out each word, find its meaning in the dictionary and diagram every sentence. This strikes me as a gross misrepresentation of mechanical reasoning, and I think the crux of this misrepresentation is a confusion between reasoning and representation. Fortunately, there are other philosophers who appreciate that these are distinct issues; but they don't seem to attract as much attention as Dreyfus. One last jab in parting: However, a computer cannot recognize emotions such as anger in facial expressions, because we know of no way to break down anger into elementary symbols. Therefore, logic machines cannot see the similarity between two faces that are angry. Yet human beings can discern the similarly almost instantly. This strikes me as another example of sloppy thinking. Are we talking about a GEDANKEN experiment here? If so, how are we to define it? Are we looking at faces out of context in an attempt to infer emotion? If so, then I would claim that humans are nowhere near as good as is claimed. Indeed, man has been notorious for misreading emotion. The lack of this skill has probably perpetrated many major historical events. Seymour Papert used to accuse Dreyfus of committing the "superhuman human" fallacy by assuming that an artrificial intelligence would surpass a human one. Here is a situation where Dreyfus hasd gone out on a limb which he should have left alone. Our understanding of how PEOPLE exhibit and perceive emotion is sufficiently weak that, for the most part, artificial intelligence seems to have to good sense to leave it in peace. ------------------------------ Date: 22 Oct 87 14:21:54 GMT From: PT.CS.CMU.EDU!SPICE.CS.CMU.EDU!spe@cs.rochester.edu (Sean Engelson) Subject: The success of AI (misunderstandings) A couple of clarifications in response to recent posts: (a) My name is Engelson---NOT Engleson. (b) I did not state that we could simulate the human body and brain at this point in time. However, we could at some point, presumably, get to the point where we know precisely how the body is constructed, and construct a simulation of the physical processes that occur. This is reasonable because the human body is finite in extent, and thus there is a finite amount of information to discover, thus it can be discovered in finite (although possibly very large) time. This is why I say that computers are not a less-powerful model of computation than the human brain, as the one can simulate the other. By 'as powerful' I mean that the same computations may be performed by both; in the same sense that a serial computer is as powerful as a parallel one, as the one can simulate the other, although with a great loss of efficiency. (c) No, it would not be neccesary to simulate the physical world in our hypothetical super-computer. We could simulate the actions of the sensory inputs by filtering such things as movie-camera output, tactile sensors, etc., through a simulation of human sensory organs. We know that that is theoretically possible through the same line of reasoning as above. -Sean- -- Sean Philip Engelson I have no opinions. Carnegie-Mellon University Therefore my employer is mine. Computer Science Department ---------------------------------------------------------------------- ARPA: spe@spice.cs.cmu.edu UUCP: {harvard | seismo | ucbvax}!spice.cs.cmu.edu!spe ------------------------------ Date: 23 Oct 87 09:00:36 GMT From: mcvax!varol@uunet.uu.net (Varol Akman) Subject: Re: The success of AI (misunderstandings) In article <213@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: > > .................... > >discovered in finite (although possibly very large) time. This is why >I say that computers are not a less-powerful model of computation than >the human brain, as the one can simulate the other. By 'as powerful' > --------------------------------- Congratulations, when are you going to receive your Nobel prize for discovering that? Varol Akman, CWI, Amsterdam What is an individual? A very good question. So good, in fact, that we should not try to answer it. - DANA SCOTT ------------------------------ Date: 23 Oct 87 17:59:07 GMT From: uwslh!lishka@speedy.wisc.edu (Christopher Lishka) Subject: Re: The success of AI (misunderstandings) In article <213@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: > >A couple of clarifications in response to recent posts: > >(b) I did not state that we could simulate the human body and brain at >this point in time. However, we could at some point, presumably, get >to the point where we know precisely how the body is constructed, and >construct a simulation of the physical processes that occur. This is >reasonable because the human body is finite in extent, and thus there >is a finite amount of information to discover, thus it can be >discovered in finite (although possibly very large) time. This is why >I say that computers are not a less-powerful model of computation than >the human brain, as the one can simulate the other. By 'as powerful' >I mean that the same computations may be performed by both; in the >same sense that a serial computer is as powerful as a parallel one, as >the one can simulate the other, although with a great loss of efficiency. > I have some questions of Mr. Engelson (forgive me is I misspelled your name in my last posting), that others on the net might answer also: How do we know that a computer and a human are "as powerful" as each other? How do we know that the same computations can be performed on each "entity?" Referring back to the biological sciences (esp. Neurobiology), it would seem that there is so much that is *not* known that coming to conclusions about abstract things such as how a human body computes (especially billions of computations that we are not aware of) is a bit naive at this point. It seems like so many mistakes that were made in the past about the human body and mind: the brain as complex plumbing, the brain as a rather large telphone network, etc. Can the assumption that the two are equal in their power to compute really be made based on what humans know (and do not know) about their own functioning? Just a thought (maybe I am looking at this the wrong way...). By the same reasoning as above, is the analogy between serial and parallel computers (and a computer and human body) really a good one? The differences between any computer and a human body (based on the little we do know) is staggering. In theory, things appear to be the same. But computers do not have hormones, neurotransmitters, internal messengers, complex channels, etc. for each of their "basic" constituents (which I am assuming are cells). Now, theoretically they may not be necessary. In constructing a model, it is easy to overlook what can be implemented and what is easy to implement. But practically the mechanisms may be necessary. I don't know. No one else knows. But I do know that my Professor of Neurobiology (whom I think is a good source) as well as the Grad. Students I have spoken with *all* warn me to beware of these oversights, because the small details are what do make the difference. If these messenger molecules and different neurotransmitters and sodium/potassium/calcium channels and electrical vs. chemical channels were totally useless, why have they survived millions of years of evolution? Are we then only super-parallel processors when compared to parallel-processing computers, just as parallel-processing computers are to serial computers? >(c) No, it would not be neccesary to simulate the physical world in >our hypothetical super-computer. We could simulate the actions of the >sensory inputs by filtering such things as movie-camera output, >tactile sensors, etc., through a simulation of human sensory organs. >We know that that is theoretically possible through the same line of >reasoning as above. Is this reasonable? Could we raise a human being properly be hooking his retinal receptors to wires, his aural receptors to wires, his tongue connections to a computer simulation, etc.? Would we get a *normal* person? Personally, I don't think so, but then I don't know; noone knows. And until someone such as Hitler comes along, the question will probably remain unanswered. Now, I feel this applies to computers because we would, in effect, be doing the same thing (given that we could artificially create a model of a human in a computer). You would still need to simulate the real world in the images that you gave the machine. The images would need to respond to the machine. When the machine wanted to move, all of the images and artificial senses would need to reflect that. When the machine tried wanted to ask a question while standing on its head, twiddling it fingers, chewing gum, and computing pi to the fourth power, could the images and artificial senses fed to it effectively simulate that? (I know, it probably wouldn't have a head or do those things, so just insert any funny little thing that a "child" computer-modelled human would do at once.) Again, no small feat. Is this really possible in the future? >Sean Philip Engelson I have no opinions. Just some thoughts of mine (the above are NOT intended to be flames). I feel is a very interesting discussion, but in the end hinges on one's personal beliefs and philosophies (but then, what doesn't ;-) The usual disclaimer applies (including the bit about the cockatiels). -Chris -- Chris Lishka /lishka@uwslh.uucp Wisconsin State Lab of Hygiene <-lishka%uwslh.uucp@rsch.wisc.edu "What, me, serious? Get real!" \{seismo, harvard,topaz,...}!uwvax!uwslh!lishka ------------------------------ Date: 23 Oct 87 16:32:24 GMT From: violet.berkeley.edu!ed298-ak@jade.Berkeley.EDU (Edouard Lagache) Subject: Clarifying Dreyfus's work (Re: The Success of AI). I would like to clarify some of the aspects of Hubert Dreyfus's work that were overlooked by Stephen Smoliar in his note. I won't try to defend Dreyfus, since I doubt that many people on this SIG is really open-minded enough to consider the alternative Dreyfus proposes, but for the sake of correctness: Most of Mr. Smoliar points are in fact dealt with in his first book. His second book was intended more for the general public, thus it glosses over a number of important arguments that are in the first book. As a matter of opinion, I like the first book better, although it is probably important to read both books to understand his full position. The first book is: What Computers Can't Do, The Limits of Artificial intelligence, Harper and Row, 1979. One point where Mr. Smoliar misses Dreyfus completely is in his assumption that Dreyfus is taking about models. Dreyfus is far more radical than that. He believes that humans don't make models, rather they carry a collection of specific cases (images?) Anyone who is at all honest in this field has to admit that there are a lot of failures to be accounted for. While I feel that Dreyfus is too pessimistic in his outlook, I feel that there is value in looking at his perspective. I would hope that by reflecting on (and reacting against) such skepticism, A.I. researchers would be able to sharpen their understanding of both human and Artificial Intelligence. Edouard Lagache lagache@violet.berkeley.edu ------------------------------ End of AIList Digest ******************** 25-Oct-87 22:40:58-PST,16224;000000000001 Mail-From: LAWS created at 25-Oct-87 22:26:16 Date: Sun 25 Oct 1987 22:24-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #246 - Seminars, Conferences To: AIList@SRI.COM AIList Digest Monday, 26 Oct 1987 Volume 5 : Issue 246 Today's Topics: Seminars - Genetic Algorithms and Pallet Loading (BBN) & Applying AI to Instruction (San Diego SIGART) & Implementing Theorem Provers in Logic (UPenn), Conferences - Expert Systems in Business and Finance & Expert Systems in Agriculture & Distributed AI Workshop ---------------------------------------------------------------------- Date: Fri 23 Oct 87 16:01:14-EDT From: DDAVIS@G.BBN.COM Subject: Seminar - Genetic Algorithms and Pallet Loading (BBN) BBN Science Development Program AI/Education Seminar Series "GENETIC ALGORITHMS AND PALLET LOADING" Pat Prosser Dept. of Computer Science Univ. of Strathclyde Glasgow G1 1XH U.K. BBN Laboratories Inc. 10 Moulton Street Large Conference Room, 2nd Floor 10:30 a.m., Tuesday, November 10, 1987 Abstract: Genetic Algorithms (GA) are search techniques based on the paradigm of population genetics. A highly constrained loading problem, the loading of stacks of plates onto pallets, is used as a vehicle for measuring the suitability of a GA approach to solving sequencing problems. In this seminar the following points will be addressed: - A brief description of the generic GA - A description of the pallet loading problem and two approaches taken in solving the problem - The genetic representation used - The two genetic algorithmic solutions implemented GA1 and GA2 - Implementation and Performance issues ------------------------------ Date: 22 Oct 87 11:40:00 EDT From: "GAIL SLEMON 455-1330" Reply-to: "GAIL SLEMON 455-1330" Subject: Seminar - Applying AI to Instruction (San Diego SIGART) The San Diego chapter of ACM SIGART is meeting on November 19, Thursday evening at the University of California, San Diego, Peterson Hall, Room 103, 6:30 - 8:30 pm. Everyone is invited to attend. Free admission. APPLYING AI TO INSTRUCTION Dr. Greg Kearsley of Park Row Software This session will focus on the two major applications of Artificial Intelligence techniques to instruction: intelligent tutoring systems and expert systems. The major types of intelligent tutors will be described, along wih a discussion of design and development methodology. -- For more info on SDSIGART and the above, contact: Gail Slemon at (619) 455-1330 or GSLEMON@afhrl.arpa ------------------------------ Date: Thu, 22 Oct 87 19:07:41 EDT From: dale@linc.cis.upenn.edu (Dale Miller) Subject: Seminar - Implementing Theorem Provers in Logic (UPenn) Implementing Theorem Provers in Logic Programming Dissertation Proposal Amy Felty (felty@linc.cis.upenn.edu) Computer and Information Science University of Pennsylvania Logic programming languages have many characteristics that indicate that they should serve as good implementation languages for theorem provers. For example, they are based on search and unification which are also fundamental to theorem proving. We show how an extended logic programming language can be used to implement theorem provers and other aspects of proof systems for a variety of logics. In this language first-order terms are replaced with simply-typed lambda-terms, and thus unification becomes higher-order unification. Also, implication and universal quantification are allowed in goals. We illustrate that inference rules can be very naturally specified, and that the search operations based on this language correspond to those needed for searching for proofs. We argue on several levels that this extended logic programming language provides a very suitable environment for implementing tactic style theorem provers. Such theorem provers provide extensive capabilities for integrating techniques for automated theorem proving into an interactive proof environment. We are also concerned with representing proofs as objects. We illustrate how such objects can be constructed and manipulated in the logic programming setting. Finally, we propose extensions to tactic style theorem provers in working toward the goal of developing an interactive theorem proving environment that provides a user with many tools and techniques for building and manipulating proofs, and that integrates sophisticated capabilites for automated proof discovery. Many of the theorem provers we present have been implemented in the higher-order logic programming language Lambda Prolog. Date: Friday November 6, 1987 Location: 554 Moore Time: 1:30 PM Committee: Val Breazu-Tannen Robert Constable Jean Gallier (Chair) Andre Scedrov Advisor: Dale Miller ------------------------------ Date: Fri 23 Oct 87 10:07:43-EDT From: John C. Akbari Subject: Conference - Expert Systems in Business and Finance the first annual conference on expert systems in business and finance esib-87 [informal announcement] what: applications-oriented conference on the use of AI and expert systems in financial domains where: penta hotel, new york city (7th avenue at 33rd st.) when: 10 - 12 november 1987 how much: $525 after 9 october paper sessions: trading with AI expert systems: opportunities and issues tools and techniques for financial epxert systems financial applications business applications I business applications II business applications III strategic issues in AI development knowledge engineering challenges panel discussions: business and finance: viewing expert systems applications from the user's requirements perspectives applying AI in the real world corporate america: viewng AI technology and organizational issues from the academic perspective financial expert systems on PCs: strategies for successful implementation & integration expert systems in the business curriculum views from the press tutorials: building expert systems desktop AI: productivity and power in finance effective knowledge engineering sponsor: _the international journal of knowledge engineering_ for further info: learned information 143 old marlton pike medford new jersey 08055 USA tele. 609.654.6266 personal comments: looks like an interesting program covering a broad range of work in the financial services industry. there are several papers by groups working on the street on internal, proprietary systems. there are also several papers by vendors (syntelligence, inference, etc.) who have experience in building these systems. my opinion is that there the ratio of academic types to "real world" types is still too high, but is better than most of those expensive "conferences" put on by the market research and consulting groups. this conference should be less content-free than most. it will be interesting to see the effects of this week's crash on future efforts! ------------------------------ Date: Sat, 24 Oct 87 15:15 EDT From: Thieme@BCO-MULTICS.ARPA Subject: Conference - Expert Systems in Agriculture CALL FOR PAPERS TITLE: Integration of Expert Systems with Conventional Problem Solving Techniques in Agriculture SPONSORED BY: AAAI Applied Workshop Series DESCRIPTION: Problem solving techniques such as modelling, simulation, optimization, and network analysis have been used for several years to help agricultural scientists and practitioners understand and work with biological problems. By their nature, most of those problems are difficult to define quantita- tively. In addition many of the models and simulations that have been developed are not "user-friendly" enough to entice practitioners to use them. As a result, several scientists are integrating expert system technology with conventional problem solving techniques in order to increase robustness of their systems as well to increase usability and to aid in result interpretation. The goal of this workshop is to investigate the similarities and differences of leading scientists' approaches and to develop guidelines for similar work in the future. CONDITIONS OF PARTICIPATION: Primary authors (presumably primary investigators) of submitted manuscripts will be invited to participate in the workshop if their manuscript is selected. Manuscripts will be submitted in full six weeks prior to the workshop. The manuscripts will be reviewed for originality and clear presentation of the topic of integration by a committee appointed by the coordinating committee. Only 40 participants will be selected in order to maximize free exchange of ideas. The manuscripts will be distributed to the participants prior to the workshop in order to help them prepare questions for other authors. If there is interest, the proceedings will be published. LOCATION AND TIME: April 13-15, 1987 at the Menger Hotel in San Antonio, TX FOR MORE INFORMATION CONTACT: Dr. A. Dale Whittaker (409) 845-8379 Agricultural Engineering Department Texas A&M University WHITTAK at TAMAGEN.BITNET College Station, TX 77843-2117 Dr. Ronald H. Thieme (617) 671-3772 Honeywell Bull, Inc. 300 Concord Road THIEME at BCO-MULTICS.ARPA Mail Station 895A Billerica, MA 01821 Dr. James McKinion (601) 323-2230 Crop Science Research Laboratory Crop Simulation Research Unit P.O. Box 5367 Mississippi State, MI 39762-5367 Earl Kline (409) 845-3693 Agricultural Engineering Department Texas A&M University KLINE at TAMAGEN.BITNET College Station, TX 77843-2117 ------------------------------ Date: Wed, 21 Oct 87 23:19:02 PDT From: gasser%pollux.usc.edu@oberon.USC.EDU (Les Gasser) Subject: Conference - DAI Workshop Announcement WORKSHOP ANNOUNCEMENT - CALL FOR PARTICIPATION 8th Workshop on Distributed Artificial Intelligence Lake Arrowhead Conference Center Lake Arrowhead, CA. May 22-25, 1988 The 8th Distributed AI Workshop will address the problems of coordinated action and problem-solving among reasonably sophisticated, intelligent computational "agents." The focus will be be synthetic and pragmatic, investigating how we can integrate theoretical and experimental ideas about knowledge, planning, negotiation, action, etc. in multi-agent domains, to build working interacting agents. Participation is by invitation only. To participate, please submit an extended abstract (5-7 double-spaced pages, hard copy only) describing original work in DAI to the workshop organizer at the address below. Preference will be given to work addressing basic research issues in DAI such as those outlined below. A small number of "interested observers" will also be invited. If you are interested in being an observer, please submit a written request to attend (hard copy), with some justification. Participation will be limited to approximately 35 people. A number of submitted papers will be selected for full presentation, critique, and discussion. Other participants will be able to make brief presentations of their work in less formal sessions. There will be ample time allowed for informal discussion. All participants should plan to submit a full paper version in advance, for distribution at the workshop. Suggested topics include (but are not necessarily limited to): Describing, decomposing, and allocating problems among a collection of intelligent agents, including resource allocation, setting up communication, dynamic allocation, etc. Assuring coherent, coordinated interaction among intelligent agents, including allocating control, determining coherence, organization processes, the role of communication in coherence, plan synchronization, etc. Reasoning about other agents, the world, and the state of the coordinated process, including plan recognition, prospective reasoning, knowledge and belief models, representation techniques, domain or situation specific examples, etc. Recognizing and resolving disparities in viewpoints, representations, knowledge, goals, etc. (including dealing with incomplete, inconsistent, and representationally incompatible knowledge) using techniques such as communication, negotiation, conflict resolution, compromise, deal enforcement, specialization, credibility assessment, etc. Problems of language and communication, including interaction languages and protocols, reasoning about communication acts inter-agent dialogue coherence, etc. Epistemological problems such as joint concept formation, mutual knowledge, situation assessment with different frames of reference, etc. Practical architectures for and real experiences with building interacting intelligent agents or distributed AI systems. Appropriate methodologies, evaluation criteria, and techniques for DAI research, including comparability of results, basic assumptions, useful concepts, canonical problems, etc. For this DAI workshop, we specifically discourage the submission of papers on issues such as programming language level concurrency, fine-grained parallelism, concurrent hardware architectures, or low-level "connectionist" approaches. Please direct inquiries to the workshop organizer at the address below. ---------------------------------------------------------------- DATES: Deadline for submission of extended abstracts: February 15, 1988 Notification of acceptance: March 21, 1988 Full papers due (for distribution at the workshop): April 25, 1988 ---------------------------------------------------------------- WORKSHOP ORGANIZER: Les Gasser Distributed AI Group Computer Science Department University of Southern California Los Angeles, CA. 90089-0782 Telephone: (213) 743-7794 Internet: gasser@usc-cse.usc.edu ---------------------------------------------------------------- WORKSHOP PLANNING COMMITTEE: Miro Benda (Boeing AI Center) Phil Cohen (SRI) Lee Erman (Teknowledge) Michael Fehling (Rockwell) Mike Genesereth (Stanford) Mike Georgeff (SRI) Carl Hewitt (MIT) Mike Huhns (MCC) Victor Lesser (UMASS) N.S. Sridharan (FMC Corp) ---------------------------------------------------------------- Support for this workshop and for partial subsidy of participants' expenses has been provided by AAAI; other support is pending. ------------------------------ End of AIList Digest ******************** 25-Oct-87 22:47:25-PST,14567;000000000000 Mail-From: LAWS created at 25-Oct-87 22:33:51 Date: Sun 25 Oct 1987 22:30-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #247 - Knowledge, Neural Terminology, Design, Linguistics To: AIList@SRI.COM AIList Digest Monday, 26 Oct 1987 Volume 5 : Issue 247 Today's Topics: Clarification - Knowledge Soup, Neuromorphic Systems - Historical Terminology, Bibliographies - Design Automation/Assistance & Computational Linguistics ---------------------------------------------------------------------- Date: 24 October 1987, 20:03:38 EDT From: john Sowa Subject: Knowledge Soup An abstract of a recent talk I gave found its way to the AIList, V5 #241. But along the way, the first five sentences were lost. Those sentences made a distinction that was at least as important as the rest of the abstract: Much of the knowledge in people's heads is inconsistent. Some of it may be represented in symbolic or propositional form, but a lot of it or perhaps even most of it is stored in image-like forms. And some knowledge is stored in vague "gut feel" or intuitive forms that are almost never verbalized. The term "knowledge base" sounds too precise and organized to reflect the enormous complexity of what people have in their heads. A better term is "knowledge soup." Whoever truncated the abstract also changed the title "Crystallizing theories out of Knowledge Soup" by adding "(knowledge base)". That parenthetical addition blurred the distinction between the informal, disorganized knowledge in the head and the formalized knowledge bases that are required by AI systems. Some of the most active research in AI today is directed towards handling that soup and managing it within the confines of digital systems: fuzzy logic, various forms of default and nonmonotonic reasoning, truth maintenance systems, connectionism and various statistical approaches, and Hewitt's due-process reasoning between competing agents with different points of view. Winograd and Flores' flight into phenomenology and hermeneutics is based on a recognition of the complexity of the knowledge soup. But instead of looking for ways of dealing with it in AI terms, they gave up. Although I sympathize with their suggestion that we use computers to help people communicate better with each other, I believe that variations of current AI techniques can support semi-automated tools for knowledge acquisition from the soup. More invention may be needed for fully automated systems that can extract theories without human guidance. But there is no clear evidence that the task is impossible. ------------------------------ Date: Sat, 24 Oct 1987 14:44 EDT From: MINSKY%OZ.AI.MIT.EDU@XX.LCS.MIT.EDU Subject: AIList V5 #244 Neuromorphic Terminology Terms like "neural networks" were in general use in the 1940's. To see its various forms I suggest looking through the Bulletin of Mathematical Biophysics in those years. For example, there is a 1943 paper by Landahl, McCulloch and Pitts called "A statistical consequence of the logical calculus of Nervous Nets" and a 1945 paper by McCulloch and Pitts called "A heterarchy of values determined by the topology of Nervous Nets. It is true that Papert and I confused this with the title of another McCulloch Pitts 1943 paper, which used the term "nervous activity" instead. Both papers were published together in the same journal issue. In any case, "neural networks" and "nervous nets" were already in the current jargon. In the original of my 1954 thesis, I called them "Neural-Analog Networks, evidently being a little cautious. But in the same year I retitled it for publication (for University Microfilms) as "Neural Nets and the Brain Model Problem". My own copy has "Neural Netorks and the ..." printed on its cover. My recollection is that we all called them, simply, "neural nets". A paper of Leo Verbeek has "neuronal nets" in its title; a paper of Grey Walter used "Networks of Neurons"; Ashby had a 1950 paper about "randomly assembled nerve networks. Farley and Clark wrote about "networks of neuron-like elements". S.C.Kleene's great 1956 paper on regular expressions was entitled "Representation of events in Nerve Nets and Finite Automata". Should we continue to use the term? As Korzybski said, the map is not the world. When a neurologist invents a theory of how brains learn, and calls THAT a neural network, and complains that other theories are not entitled to use that word, well, there is a problem. For even a "correct" theory would apply only to some certain type of neural network. Probably we shall eventually find that there are many different kinds of biological neurons. Some of them, no doubt, will behave functionally very much like AND gates and OR gates; others will behave like McCulloch-Pitts linear threshold units; yet others will work very much like Rosenblatt's simplest perceptrons; others will participate in various other forms of back-propagated reinforcement, e.g., Hebb synapses; and so forth. In any case we need a generic term for all this. One might prefer one like "connectionist network" that does not appear to assert that we know the final truth about neurons. But I don't see that as an emergency, and "connectionist" seems too cumbersome. (Incidentally, we used to call them "connectionistic" - and that has condensed to "connectionist" for short.) ------------------------------ Date: Sat 24 Oct EDT 1987 05:46 From: frodo%research.att.com@RELAY.CS.NET Subject: Re: AI & Design Automation, Design Assistance A few references on AI and VLSI CAD %author Kowalski, T. J. %title An Artificial Intelligence Approach to VLSI Design %publisher Kluwer %address Boston, MA %date 1985 %keyword DAA %author Wolf, W. H. %author Kowalski, T. J. %author McFarland, M. C. %title Knowledge Engineering Issues in VLSI Synthesis %journal Proceedings of the National Conference on Artificial Intelligence %pages 866-871 %date 1986 %author Kowalski, T. J. %author Geiger, D. J. %author Wolf, W. H. %author Fichtner, W. %title The VLSI Design Automation Assistant: From Algorithms To Silicon %journal Design and Test of Computers %volume 2 %number 4 %pages 33-43 %date August, 1985 %author McFarland, M. C. S.J. %author Kowalski, T. J. %title Assisting DAA: The Use of Global Analysis in an Expert System %journal Proceedings of the IEEE International Conference on Computer Design %pages 482-485 %publisher IEEE %address New York, NY %date October 6, 1986 %keyword DAA BUD ------------------------------ Date: 21 Oct 87 20:12:35 GMT From: russell!goldberg@labrea.stanford.edu (Jeffrey Goldberg) Subject: Computational Linguistics Bibliography by E-Mail (CLBIB) It is possible to do a keyword search on a > 1700 entry bibliography of work in computational linguistics published in the 1980's. Here is how: Computational Linguistics & Natural Language Processing Bibliography by Mail There is a large (> 1700 items) bibliography of 1980s natural language processing and computational linguistics sitting on a Sun called Russell at CSLI. Anyone with a computer account can now search this bibliography and get a listing of the result by using electronic mail. INSTRUCTIONS The keywords used for the lookup are to be given in the subject line of your mail message addressed to clbib@russell.stanford.edu (36.9.0.9). The body of your message will be thrown away. Here is an example: % mail clbib@Russell.Stanford.EDU Subject: Woods ATN 1980 . EOT Null message body; hope that's okay % Or more compactly: % Mail -s "woods atn 1980" clbib@Russell.Stanford.EDU < /dev/null And here is what you would receive in return: >>> Date: Wed, 11 Jul 87 12:03:35 PST >>> To: yourname >>> Subject: CLBIB search: Woods ATN ... %A T.P. Kehler %A R.C. Woods %T ATN grammar modeling in applied linguistics %D 1980 %P 123-126 %J ACL Proceedings, 18th Annual Meeting %A William A. Woods %T Cascaded ATN grammars %D 1980 %V 6 %N 1 %P 1-12 %J American Journal of Computational Linguistics This example show one mailing from a Unix machine, but you can mail CLBIB from any machine and get a result, provided you remember to put your search keys in the "Subject:" field of the message. The entries you get are in standard Unix 'refer' format (see the man page). You may put between one and eight keywords in the mail "Subject: " field, and each keyword can be any string of characters (name, date, topic, etc.) that you think likely to be found in the items of interest (case is ignored). The list of keywords is interpreted conjunctively: "Woods" gets you everything published by anyone called "Woods" in the 1980s, whereas "Woods 1983" narrows that down to just the 1983 papers (or papers whose first or last page number is "1983") by persons named "Woods" (or whose title refers to "woods"), and, of course, there may be no such items (so the reply would contain nothing). Only the first six characters in a keyword are significant, so "generation" is indistinguishable from "generalized", and "Anderson" is indistinguishable from "Andersson". You should bear this in mind when you consider the relevance of what you receive to your intended request. To take up less CPU at this end, please use as your first keyword the one that will narrow selections down the most. The first key may not be a year. If the first key is "help", you will be sent this file. BUGS The system is no better than the mail connections. This system is worse than the mail connections. The return address is determined only from information in the "From" field. "Reply-To:" should be checked but it is not. The return parsing is stupid and doesn't know all there is to know about RFC822 mail headers. The "From" and "Subject" fields must have exactly the "F" and the "S" in uppercase. It is impossible to seach for only the item "help". (You get this file if the first key on a subject line is "help") It is impossible to get all of the entries for one year. [This is not a bug. If you want the entire list you can follow the instructions about such things below.] The mail handling scripts were written by linguists, not by programmers. The scripts are fragile and the system may be taken down without notice at anytime. THE BIBLIOGRAPHY Some sense of the scope of the bibliography can be gathered from the following summary information. Here are the authors who find themselves with a dozen or more of their 1980s publications included: 25 Aravind K. Joshi 19 Bonnie Lynn Webber 18 Robert C. Berwick 18 Jaime G. Carbonell 17 David D. McDonald 15 Philip J. Hayes 15 Wendy G. Lehnert 15 Fernando C.N. Pereira 14 Kathleen R. McKeown 14 Karen Sparck-Jones 13 Eugene Charniak 13 Barbara J. Grosz 13 Jerry R. Hobbs 13 Martin Kay 13 Stuart M. Shieber 12 Douglas E. Appelt 12 Philip R. Cohen 12 C. Raymond Perrault 12 Graeme D. Ritchie 12 Ralph M. Weischedel 12 Yorick A. Wilks And the papers included distribute across the years like this: 1980: 207 1981: 138 1982: 211 1983: 240 1984: 219 1985: 247 1986: 353 1987: 117 The 1987 figure includes the contents of this year's ACL Proceedings, and the relevant papers in AAAI-87, but not those from the upcoming IJCAI meeting in August nor the as-yet-unpublished 1987 European ACL Proceedings. Machine-readable copies of the entire bibliography are available on standard MS-DOS 360K DS/DD disks. Write to Ms Sheila Lee, CSRP Series, School of Cognitive Sciences, University of Sussex, BRIGHTON BN1 9QN, UK, asking for a copy of the CL-NLP8X.BIB bibliography disk, and enclose a check for $16.00 to cover media, handling, packing and postage costs. A hardcopy version of the entire bibliography with a permuted index of titles and an index to nonprimary authors is to be published by CSLI/Chicago University Press in November 1987 - details below: %A Gerald Gazdar %A Alex Franz %A Karen Osborne %A Roger Evans %D 1987 - in press %T Natural Language Processing in the 1980's - A Bibliography %C Stanford %S CSLI Lecture Notes %I Chicago University Press If there is a problem with this program please send a note to: clbib-request@Russell.stanford.edu But only questions about the mailing system can be dealt with. Problems with the content of the bibliography (typos, omissions, etc) are not something that we are capable of coping with here. SEE ALSO refer(1) Mail(1) tib(local) AUTHORS & ACKNOWLEDGEMENTS The bibliography was compiled at the University of Sussex under the direction of Gerald Gazdar by Gerald Gazdar, Alex Franz, Karen Osborne, and Roger Evans. Initial c-shell scripts were written by Evans and Gazdar at Sussex. They were overhauled by Jeff Goldberg at CSLI. In addition to more standard Unix tools (awk(1), sed(1), Mail(1), etc), refer(1) (available on most Unix distributions) and Tib (available on the Unix TeX distribution) are employed. Unix is a trade mark of AT&T. SUMMARY To search bibliography mail to clbib@Russell.stanford.edu with the keywords for the search as your Subject line. To get a help file send to clbib@Russell.stanford.edu with "help" as the first keyword in your subject line. To get in touch with real people, send to clbib-request@Russell.stanford.edu Information about getting a hardcopy of the bibliography with indicies will be forthcoming any day now. -- Jeff Goldberg ARPA goldberg@russell.stanford.edu UUCP ...!ucbvax!russell.stanford.edu!goldberg ------------------------------ End of AIList Digest ******************** 25-Oct-87 22:53:56-PST,22217;000000000001 Mail-From: LAWS created at 25-Oct-87 22:47:10 Date: Sun 25 Oct 1987 22:41-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #248 - OCR, Introductory Prolog, Flawed Minds To: AIList@SRI.COM AIList Digest Monday, 26 Oct 1987 Volume 5 : Issue 248 Today's Topics: Query - Character Recognition, Tools - OCR & Character Recognition, Education - Introductory Prolog, Comments - Flawed Minds & Goal of AI ---------------------------------------------------------------------- Date: 16 Oct 87 12:47:55 GMT From: mcvax!ukc!its63b!hwcs!zen!vic@uunet.uu.net (Victor Gavin) Subject: Character recognition I have been puttering about for the past few weeks with an HP ScanJet (one of those 300dpi digitizers). I have been asked to write some software which can (given an image produced by the scanner) reproduce the original text of the paper in a machine readable form. The text will normally be numbers and the image will initially be a bit pattern. If someone can point me to some introductory texts on character recognition I would be grateful. If someone has already tackled this problem, any help I can get will be much appreciated. vic -- Victor Gavin Zengrange Limited vic@zen.co.uk Greenfield Road ..!mcvax!ukc!zen.co.uk!vic Leeds LS9 8DB +44 532 489048 England ------------------------------ Date: 24 Oct 87 21:14:14 GMT From: phri!roy@NYU.EDU (Roy Smith) Subject: Re: Character recognition In article <641@zen.UUCP> vic@zen.UUCP (Victor Gavin) writes: > I have been asked to write some software which can (given an image > produced by the scanner) reproduce the original text of the paper in a > machine readable form. I don't know much about it, but a company called DEST markets a 300-dpi scanner for the Macintosh (and, I think, IBM-PC) for about $2k, including character recognition software. Unless your application has some special requirements, I would imagine getting one of these jobs would be a lot more cost-effective than writing your own software. I've added comp.sys.mac to the Newsgroups line to see if anybody there has any experience with the DEST they could share. While I'm at it, can somebody compare and contrast the O($2k) scanners with the el-cheapo Thunderscan for me. What to the "real" scanners have going for them that I can't do with a Thunderscan? -- Roy Smith, {allegra,cmcl2,philabs}!phri!roy System Administrator, Public Health Research Institute 455 First Avenue, New York, NY 10016 ------------------------------ Date: 25 Oct 87 00:51:32 GMT From: dewey.soe.berkeley.edu!oster@ucbvax.Berkeley.EDU (David Phillip Oster) Subject: Re: Character recognition In article <2984@phri.UUCP> roy@phri.UUCP (Roy Smith) writes: >In article <641@zen.UUCP> vic@zen.UUCP (Victor Gavin) writes: >> from a scanner image reproduce the original text of the paper in a >> machine readable form. >can somebody compare and contrast the O($2k) scanners with the el-cheapo >Thunderscan for me. What to the "real" scanners have going for them that I >can't do with a Thunderscan? Thunderscan offers very high quality scanning, at resolutions up to 300 dpi, and up to 5 bits per pixel. (32 grays.) It can handle originals up to 15" wide (in a wide carriage imagewriter) and at least 32767 scan lines long. (I haven't actually tried anything longer than 11", but when it finishes, the "continue scan" button is still waiting to be presssed.) However, it is slow, (5 to 40 minutes, depending on resolution and size of original.) and only works on single sheet, thin, bendable material. (The material has to fit in the imagewriter printer.) That means you'd do well to have a xerographic copier handy. The expensive scanners are flat bed, copier style machines, and do their work faster (can't be too much faster, though. It takes 15minutes to send an 8"x10" page at 1-bit per pixel 300dpi, over a 9600 baud line if you do not use a compressing transfer protocol.) Olduvai Software makes a line of software that parses scanned pages back into text. Either the current issue of MacUser has a review, or I saw it in a recent copy of MacWeek, but for < $200.00 you get a software package to do syntactic pattern recognition of letter features, to determine the ASCII for the scanned page. It is still cheaper to hire a human typist, but soon the cost balance will flip the other way. (I expect that copy shops will offer a service: bring in your books and blank disks, and for a few cents a page, get them digitized to ASCII. (And won't that boost our needs for on-line storage (What, only 300Gigabytes! How do your get by with such a small library?))) (note, I've directed followups to just comp.misc. If people want to continue this discussion, they can read it there.) --- David Phillip Oster --A Sun 3/60 makes a poor Macintosh II. Arpa: oster@dewey.soe.berkeley.edu --A Macintosh II makes a poor Sun 3/60. Uucp: {uwvax,decvax,ihnp4}!ucbvax!oster%dewey.soe.berkeley.edu ------------------------------ Date: 24 Oct 87 16:42:23 GMT From: unc!bts@mcnc.org (Bruce Smith) Subject: Re: Suggestions for Course Turbo Prolog for an AI course? Why not FORTRAN, for that matter? Quoting (without permission) from Alan Bundy's Catalog of AI Tools: FORTRAN is the programming language considered by many to be the natural successor of LISP and Prolog for AI research. Its advantages include 1. It is very efficient for numerical computation (many AI programs rely heavily on number crunching techniques). 2. AI programs tend to be very poorly constructed, meaning that control needs to move frequently from one part of a program to another. FORTRAN provides a special mechanism for achieving this, the so-called GOTO statement. 3. FORTRAN provides a very efficient data structure, the array, which is particularly useful if, for example, one wishes to process a collection of English sentences each of which has the same length. ______________________ Bruce T. Smith, UNC-CH bts@unc.edu ------------------------------ Date: 24 Oct 87 21:09:34 GMT From: rocky!wagner@labrea.stanford.edu (Juergen Wagner) Subject: Re: Suggestions for Course Great! I believe, Bruce hit the right point. Teaching a programming language whose conceptual structure is that different from what most people think of programming languages, should not be done using almost a counterexample of that paradigm. Some people are convinced that TurboP#$@$ is a real Prolog (which might be true in their understanding of AI languages), and there might be applications where sliding away from PASCAL over TurboProlog to (REAL) Prolog (just to introduce changes step by step), but it is definitively no good choice for teaching typical AI programming techniques which (by their nature) require symbol crunching rather than number crunching. And if I first have to write a bundle of declarations before I find out that this highly nested and flexible data structure I have in mind cannot be implemented that way, this is not what I expect of such a programming language. Sure, TurboP#$@$ is available on IBM/PCs. But there are also other nice Prologs around, even a Public Domain one (SBProlog, mentioned in this newsgroup some time ago). So, why not take a real Prolog even if it is only line-oriented, and even if you have to write the main parts of your programs outside Prolog with a conventional text editor? The idea of an AI course should be to convey to basic principles and the special way of thinking and reasoning about (so-called) AI problems. Exploring and experimenting with programs gives a good impression of that. Ok. No more flames about TurboP#$@$. Juergen Wagner, (USENET) gandalf@portia.stanford.edu Center for the Study of Language and Information (CSLI), Stanford CA ------------------------------ Date: 25 Oct 87 03:50:08 GMT From: violet.berkeley.edu!ed298-ak@jade.Berkeley.EDU (Edouard Lagache) Subject: Re: Suggestions for Course (Getting a PROLOG cheap or free) I think that PROLOG is a much better choice for an intro course on A.I. (someday maybe I will write a paper on why). As to getting a PROLOG to use on IBM PCs, there are a number of Public Domain PROLOGs around. One that I think would be fine for this use is put out by Automata Design Associates and can be found in various software libraries or contact them at: A.D.A. 1570 Arran Way Dresher, PA, 19025 (215) 646-4894 I think they charge $10 for a copy of their PD PROLOG, but that would be a one time investment. Edouard Lagache lagache@violet.berkeley.edu ------------------------------ Date: 25 Oct 87 15:44:48 GMT From: duke!gleicher@mcnc.org (Michael Gleicher) Subject: Re: Suggestions for Course (Couldn't bear it any longer - I have to put my two cents in) (I am not a teacher - I am a student who has gone through the courses in question) An excellent point has been brought up - What is the real reason for wanting to teach Prolog in an AI course? (by the way, replace prolog with lisp in most cases in this article) 1) Because you want to foster the belief that Prolog and AI go together - This is downright BAD and wrong! 2) Because in order to read much of the literature, you must understand Lisp/Prolog because it gets refered to alot - This one I agree with 3) Because it is the IN thing to do - I'm not even going to comment 4) Because it allows rapid prototyping so a small system that really solves problems can be built in a short amount of time - If this is your real goal, be sure not to get sidetracked. In the AI course that I took, we were taught prolog, and wrote programs to solve non-ai problems. I neither learned about how to write AI programs nor how to rapidly build a system in Prolog. I did learn some bad Prolog habits because I was trying to program prolog the same way that I would have coded in C - because there wasn't enough time for someone to show me to think otherwise. On the upside, We did have an assignment with DCG's that was interesting (only that had any notion of an AI problem to solve (natural language) or that showed me a place where I couldn't use conventional programming tactics) What using AI in a class DID do for me was allow me to get a summer job where I really learned prolog, doing things that were NOT AI. The best part of my AI class was when we looked briefly at many areas of interest in the current research (expert systems, natural language, planning, connectionism, ...). Unfortunately, this was at the end of the course. Maybe a more effective way to teach AI would be to show the applications and use them to justify why you need to study logic and predicate calculus and frames and ... Using Turbo prolog could only accomplish #1 and #3 on the list above, but other people have already said this. Again, I am not a teacher, just a student who has taken these courses, and is still interested in the subject DESPITE the courses. And - PROLOG IS NOT JUST FOR ``AI'' Mike ------------------------------ Date: 25 Oct 87 02:50:37 GMT From: kludge@pyr.gatech.edu (Scott Dorsey) Reply-to: kludge@pyr.gatech.edu (Scott Dorsey) Subject: Humor - Flawed Minds In article <8710240608.AA19274@ucbvax.Berkeley.EDU> RCSMPA::HAMILTON@gmr.COM ("William E. Hamilton, Jr.") writes: >Of course the human mind is flawed. The proof is quite straightforward. Fred the Football Player says, "I don't get no A's in none of my classes. Dis is either cause da professirs what give de tests is flawed, or cause I is flawed. If eider one of dese tings is true, den I can state catigorikly dat dere is at least won person what got a flawed mind. If one person got a flawed mind, then he isn't no good at judgin' if udder people also got flawed minds. Den, since no person can tell if deir mind is flawed (cause if you mind is flawed, it might be flawed in way dat make you tink it aint flawed. If it aint flawed, den you no it aint flawed, but eider way you tink da same ting, so ya gotta assume dat it's flawed), dey gotta assume dat it is, so dey aint able to be sure dat dey can tell if anyone else got one flawed mind. Derefore ya gotta assume dat everybody got a flawed mind." -- Scott Dorsey Kaptain_Kludge SnailMail: ICS Programming Lab, Georgia Tech, Box 36681, Atlanta, Georgia 30332 Internet: kludge@pyr.gatech.edu uucp: ...!{decvax,hplabs,ihnp4,linus,rutgers,seismo}!gatech!gitpyr!kludge ------------------------------ Date: 22 Oct 87 16:29:25 GMT From: mcvax!ukc!its63b!hwcs!hci!gilbert@uunet.uu.net (Gilbert Cockton) Subject: Re: What the hell does flawed mean, anyway? In article <1373@houdi.UUCP> marty1@houdi.UUCP (M.BRILLIANT) writes: > >I claim that with respect to any referent the mind is flawed. >If any reader can define any referent with respect to which the >mind is perfect, I will admit my argument is flawed. Imperfection? Pointing to one's belly button Making excuses ... ... ... ... ... -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ Date: 22 Oct 87 16:20:09 GMT From: mcvax!ukc!its63b!hwcs!hci!gilbert@uunet.uu.net (Gilbert Cockton) Subject: Re: Goal of AI: where are we going? In article <1368@houdi.UUCP> marty1@houdi.UUCP (M.BRILLIANT) writes: >Factually, we know the mind is flawed because we observe that it does >not do what we expect of it. I expect my car to fetch my shoes I observe that my car does not fetch my shoes My car is flawed. I expect my dog to not move from the fire when I come to put more coal on I observe that my dog is moving when I come to put more coal on My dog is flawed I expect the word foliage to mean any "leaves" on trees shrubs I observe that people in New England use the word to mean Autumn leaves People in New England are flawed Wow! This must be logic we're seeing :-) Now for an argument based only on my understanding of what it is to convince: We can expect nothing untoward from something we do not fully understand at the level of a predictive model. I understand my car, I do not understand dogs or New Englanders. -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ Date: 23 Oct 87 15:00:25 GMT From: mcvax!ukc!stc!idec!camcon!ijd@uunet.uu.net (Ian Dickinson) Subject: Re: Is the human mind flawed? in article <2809@sdsu.UUCP>, caasi@sdsu.UUCP (Richard Caasi) says: > > If the human mind was flawless we wouldn't be debating this issue. > To determine how flawed the human mind is we need to first define the > characteristics of a flawless or perfect mind. Any suggestions? My mind does exactly what I want it to do. I like to be emotive, to be able to intuit, guess, make mistakes and learn from them, do silly things to let off steam, laugh at obscure jokes etc. All of these abilities could be regarded as flaws in a device which aspired to mechanistic perfection. But I like my mind - for me it _is_ perfect (although maybe not so to another person). > Drawing an analogy with ideal operational amplifiers > in electronics, .... Hum. I can't think of a machine that I would like to use as an analogy here. One problem is that we know that the individual components of the brain (perhaps more analogous to an electronic device) have pretty awful performance characteristics, but the *mind* as a whole has characteristics that no machine in existence today can begin to match. So, whilst I have no doubt that we can create technology that does improve on the metrics listed in the posting (indeed I am actively involved in helping to do so), I *do* doubt that this will get us much nearer to a mindful machine. > Question: Does such a mind exist or is nothing perfect in the real > world? Ultimately, reality is all we have. End of problem. -- Ian Dickinson Cambridge Consultants Ltd, AI group (0223) 358855 [U.K.] uucp: ...!seismo!mcvax!ukc!camcon!ijd or: ijd%camcon.uucp >> Disclaimer: All opinions expressed are my own (surprise!). << >> To dance is to live, but the dance of life requires many strange steps << ------------------------------ Date: 22 Oct 87 15:51:10 GMT From: mcvax!ukc!its63b!hwcs!hci!gilbert@uunet.uu.net (Gilbert Cockton) Subject: Re: Goal of AI: where are we going? In article <15196@topaz.rutgers.edu> josh@topaz.rutgers.edu (J Storrs Hall) writes: >In Western thought it has been realized at long and arduous last that >the appeal to authority is fallacious. Tell that to the judge. This understanding of Western social practices seems weak given its confusion of intellectual idealism with social reality. Authority counts for far more than rationality or science. >Experiment works; the real world exists; Not true all the time - scientific method is flawed, as any sophomore who's studied epistemology can tell you. The modern command over nature is due, not to a slavish and unimagintive application of statistical inference and hypothetico-deductive reasoning, but to an engagement which combines rigour, rationality (self-critical candour) and imagination. This view of reality and experiment is very dated and it's time some of us ignored the off-the-cuff dogma of our chemistry and physics teachers (rarely real people :-) ) and caught up with modern Western thinking (and eternal practice). > objective standards can be applied. Even to people. They must be proved objective first though, so this argument is empty. What is an objective standard? I admit the value of the idea, otherwise our concepts of morality would be weakened. But the term is not to be used lightly. "flawed" is not an objective standard, though it can be defined idiosyncratically and after the fact to correspond to standards which are. Calling the human mind "flawed" in essence could be being motivated by a lack of fit with an AI model - now shouldn't this lack of fit suggest the model is flawed and not the human mind? Note that at the end of the day, the unimaginative application of any method is less important than the people who are convinced, and remain convinced over the rest of their life. Science and convincement are not one and the same, and it is the latter which guides human life. >It is true that most AI researchers "believe that >the mind is a machine", but it seems that the alternative is to >suggest that human intelligence has a supernatural mechanism. No, Mind is extra/para-natural - we cannot observe it as we do nature, and thus the values of science do not apply. More spiritual and humanist approaches do. By the way, as a historan originally, I would hold that humanist and spiritual views of human nature have dominated, and continue to dominate, the public thinking on Man. Reductionist mechanical scientists appear to be an ugly minority who have little *respectful* social contact outside their own self-congratulating cliques. >The anti-scientific mentality is an emotional >excuse used to avoid thinking clearly. It would be much more honest >to say "I don't want to think, it's too hard work." There are other interpretations of this. I wouldn't use, for example, predicate logic (and thus Frames, semantic nets, etc), to describe the design process, not because it is too hard, but because it becomes a cretinous tool when describing such a rich human phenomenum. Thus I am not avoiding hard work; I am avoiding *fruitless* work. Many workers in AI would do better if they stopped trying to cram the world into an impoverished computational representation and actually explored the rich range of non-computable knowledge representations (e.g. the Novel, the painting, psalms, the monographs of the liberal arts). If this is all too inaccessible to their critical abilities, they could at least read some of the established works of scholarship on semantics (e.g. Lyons' 2 volumes). >The champions of irrationality, mysticism, and superstition have >emotional problems which bias their cognitive processes. Their minds are flawed This is very sad. I think the author is missing something, somewhere. I cannot believe that those who share a same higher view of humanity are misleading themselves. What does the author's friends think? -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ End of AIList Digest ******************** 25-Oct-87 23:00:01-PST,16557;000000000000 Mail-From: LAWS created at 25-Oct-87 22:49:53 Date: Sun 25 Oct 1987 22:48-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #249 - Success of AI To: AIList@SRI.COM AIList Digest Monday, 26 Oct 1987 Volume 5 : Issue 249 Today's Topics: Comments - The Success of AI ---------------------------------------------------------------------- Date: 23 Oct 87 13:23:05 GMT From: cbmvax!snark!eric@rutgers.edu (Eric S. Raymond) Subject: Re: The Success of AI In article <1342@tulum.swatsun.UUCP>, scott@swatsun (Jay Scott) writes: >[quoting me:] >> In *each case*, these problem areas were defined out of the AI field as soon >> as they spawned halfway-usable technologies and acquired their own research >> communities. > > Here's one speculation: People see intelligence as mysterious, intrinsically > non-understandable. So anything understood can't be part of intelligence, > and can't be part of AI. I assume this was what Eric had in mind in a > previous article, when he mentioned "hidden vitalist premises". Yes, that is precisely what I intended. > Any other good ideas? Maybe :-). A friend once told me that she'd read that human institutions reach a critical size at 250 people; that that is the largest social unit for which a single member can keep a reasonable grasp on the capabilities and style of everyone else in the group. This insight explains the allegedly remarkably consistent size of pre-industrial villages in areas where enough settlement land is available so that people can move elsewhere when they want. There is supposedly one well-known company that has found that the productivity gains from holding their work units down to this size more than justify the diseconomies of scale from small plants. This idea gets some confirmation from my experience of SF fandom, a totally voluntarist subculture that has, historically, thrown off sub-communities like yeast buds (SCA, Trek fandom, the Darkovans, the Dr. Who people, etc. etc.). We even have a name for these 'buds'; they're called "fringe fandoms" and the people in them "fringefen" (the correct plural of "SF fan" is, by ancient tradition "SF fen"). In this context, the theory needs a little generalizing; what seems to count for that magic 250 is not the number of self-described "Xites", but rather the smaller number of *organizers* and *regulars*; the people that maintain the subculture's communications networks and set its style. Now: let's assume a parallel division in science between "stars" (the people who do, or are seen to be doing, the important work) and "spear carriers" (the people who fill in the corners, tie down the details, go after the last decimal places, and get most of the grants ;-)). We then have: RAYMOND'S HYPOTHESIS: A scientific field with more than 250 "stars" will tend to fragment into subspecialties more and more strongly as the size increases. It would be interesting to look at other classes of voluntarist subcultures (like, say, fringe political parties) to see if a similar pattern holds. -- Eric S. Raymond UUCP: {{seismo,ihnp4,rutgers}!cbmvax,sdcrdcf!burdvax,vu-vlsi}!snark!eric Post: 22 South Warren Avenue, Malvern, PA 19355 Phone: (215)-296-5718 ------------------------------ Date: 23 Oct 87 21:23:31 GMT From: ihnp4!chinet!nucsrl!coray@ucbvax.Berkeley.EDU (Elizabeth) Subject: Re: The success of AI (misunderstandings) in reponse to: spe@SPICE.CS.CMU.EDU (Sean Engelson) / 9:21 am Oct 22, 1987 / > This is reasonable because the human body is finite in extent, > and thus there is a finite amount of information to discover, > thus it can be discovered in finite (although possibly very large) time. I am planning on gracefully failing my qualifiers in just two weeks, and one of the questions I plan to fail will have to do with decidability. Because now I know that I will blithely point out that language is finite in extent and thus there is only a finite amount of information which it can convey, so why worry about unprovable true theorems? We'll just prove all the true ones (in possibly very large finite time?) and then see if the theorem of interest is in this finite set. Grade -2. ------------------------------ Date: Saturday, 24 October 1987, 18:41-EDT From: nick@MC.LCS.MIT.EDU Subject: The success of AI (misunderstandings) In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: >Given a sufficiently powerful computer, I could, in theory, simulate >the human body and brain to any desired degree of accuracy. You are in good company. Laplace thought much the same thing about the entire physical universe. However, some results in chaos theory appear to imply that complex real systems may not be predictable even in principle. In a dynamic system with sufficiently 'sensitive dependence on intial conditions' arbitrarily large separations can appear (in the state space) between points that were initially arbitrarily close. No conceivable system of measurement can get around the fact that the behavior of the system itself 'systematically' erodes our information about its state. For a good intro to chaos theory, see the article by Farmer, Packard, et. al. in Scientific American December 86.. ------------------------------ Date: 24 Oct 87 20:41:29 GMT From: ihnp4!homxb!whuts!mtune!codas!usfvax2!pdn!alan@ucbvax.Berkeley.E DU (Alan Lovejoy) Subject: Re: The Success of AI In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: /Given a sufficiently powerful computer, I could, in theory, simulate /the human body and brain to any desired degree of accuracy... /...if I can simulate the body in a computer, the /computer is a sufficiently powerful model of computation to model the /human mind... The ultimate in "machine emulation"!!!! Why does this remind me of Chomsky's concept of 'weak' and 'strong' equivalence between grammars? Hmmm... --alan@pdn ------------------------------ Date: 24 Oct 87 20:52:32 GMT From: ihnp4!homxb!whuts!mtune!codas!usfvax2!pdn!alan@ucbvax.Berkeley.E DU (Alan Lovejoy) Subject: Re: The Success of AI In article <224@bernina.UUCP> srp@bernina.UUCP (Scott Presnell) writes: /In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: />Given a sufficiently powerful computer, I could, in theory, simulate />the human body and brain to any desired degree of accuracy. / /Horse shit. The problem is you don't even know exactly what you are /simulating! ... /For instance, dreams, are they logical?, do they fall in a pattern?, a computer /has got to have them to be a real simulation of a body/mind, but you cannot /simulate what you cannot accurately describe. Simulated horse shit! I can write a simulator for the IBM-PC to run on a Macintosh-II, without knowing or understanding all the IBM-PC programs that will ever run on it. The same is in principle possible when the machine being emulated is a human body. /Let's get down to a specific case: /I propose that given any amount of computing power, you could not presently, /and probably will never be able to simulate me: Scott R. Presnell. /My wife can be the judge. Which wife? The one being simulated by the computer as part of the simulated environment in which you are being simulated? How would you or she know which "world" you belonged to? --alan@pdn ------------------------------ Date: 24 Oct 87 21:08:27 GMT From: ihnp4!homxb!whuts!mtune!codas!usfvax2!pdn!alan@ucbvax.Berkeley.E DU (Alan Lovejoy) Subject: Re: The Success of AI In article <1993@gryphon.CTS.COM> tsmith@gryphon.CTS.COM (Tim Smith) writes: /In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: /+===== /| Given a sufficiently powerful computer, I could, in theory, simulate /| the human body and brain to any desired degree of accuracy. This /You might, for example, claim that with a /very large number of computers, all just at the edge of the /speed boundaries dictated by the laws of physics in the most /advanced materials imaginable, you could simulate a human body /and mind--but not in real time. But the simulation would have to /be in real time, because humans live in real time, doing things /that are critically time dependent (perceiving speech, for /example). You make the invalid assumption that "simulation" means that those of us in the real universe can not distinguish the simulated object or process from the real thing. It is just as valid to deal with simulations that enable one to make accurate predictions about what would happen in the real world in some well-specified scenario, even if the simulation doesn't look anything like what is simulates in the physical sense. What matters is the logical equivalence or similarity in an abstract reality. /Similarly, humans think the way they do partially because of /their size, because of the environment they live in, because of /the speed at which they move, live, and think. If the environment of an object is simulated in addition to the object itself, one need merely synchronize the object with the simulated environment as to speed, size, etc. --alan@pdn ------------------------------ Date: 23 Oct 87 16:22:45 GMT From: mcvax!ukc!its63b!hwcs!hci!gilbert@uunet.uu.net (Gilbert Cockton) Subject: Re: The Success of AI In article <193@PT.CS.CMU.EDU> spe@spice.cs.cmu.edu (Sean Engelson) writes: >Given a sufficiently powerful computer, I could, in theory, simulate >the human body and brain to any desired degree of accuracy. This >gedanken-experiment keinen gedanken mein Herr! In **WHICH THEORY**? Cut out this use of theoretical to mean "given arbitrary fantasies". Theories have real substance, and you are obliged to elaborate on the theory before alluding to it. Given a sufficiently powerful computer, could I, in theory, get everyone on the net to like my postings? Rhetorical of course, so spare me any abusive replies :-). The point again, is that I would have to elaborate the theory and test it out to be sure. Furthermore, I could not expect everyone to be convinced, that in the event of highly unlikely (impossible I believe) universal acceptance of my postings, that my theory really was the explanation. In short, even if one dropped fantasy for science, people in general are not going to be convinced. > if I can simulate the body in a computer, the computer is a > sufficiently powerful model of computation to model the mind. Of course. Now simulate it. And of course, you won't be slowed down by reading up on all the unanswered objections to the **belief** that computable formalisms can model mind. In short, this is no contribution to the argument. >we must also accept that a computer can have a mind, if only by the >inefficient expedient of simulating a body containing a mind. Ahem. Socialisation. AI people rarely have a handle on this at all. I take it that your computer simulation of the body is going to go down to the park with you to see the ducks, go down to playgroup, start primary school and work through to a degree, mixing all the time with a wide range of people, reading books, watching TV and visiting interesting places? Look, people are people because they interact as people with people. Now, who's going to want to interact with your computer as if it were a person? Need I go on? -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ Date: 23 Oct 87 13:13:59 GMT From: mcvax!ukc!its63b!hwcs!hci!gilbert@uunet.uu.net (Gilbert Cockton) Subject: Re: The Success of AI In article <1922@gryphon.CTS.COM> tsmith@gryphon.CTS.COM (Tim Smith) writes: (the best posting on this issue I've seen) >It wasn't until computers came along that there was a >metaphor for the brain powerful enough to be taken seriously. Hence the circularity in much AI appeal to cognitive psychology. As the latter is now riddled with information processing concepts, the impulsive observer will be quick to conclude from cog. psy. research that cognition works like a computer. Wrong conclusion - many cognitive psychologists talk about mind *as if it were* a computer. Likeness, especially presumed likeness, is not the same as essence, assuming noumenal objects exist of course. >There is no reason, in principle, that a very powerful >digital computer cannot imitate a mind Apologies for picking up on this, given the writer's (deleted) qualification and probable sarcasm about arguments of this form. This may appear perverse, but what on earth are these arguments of the form "nothing in principle prevents"? They are used much by the "pure" AI misanthropes, but I can never find any substance in such arguments. Which principles? How can we argue from these principles to possibility/impossibility. After all, is there anything of any genuine interest to non-logicians which is logically impossible, rather than semantically contradictory (a married bachelor for example)? Again, I pick this up because AI zealots reach for this argument all the time, and it isn't an argument at all. (PS - no flames on "misanthrope" or "zealot", one can be studying an AI topic without losing one's humanism or one's sense of moderation. I am only characterising those who are misanthropic zealots, a specialisation and not a generalisation.) >The success rate in AI research (as well as most of cognitive >science) in the past 20 years is not very encouraging. Despite all that taxpayers' money :-) > A better concept of "mind" is what is needed now. Well said. "Better" concepts related to mind than those found in cog. sci. already exist. The starting point is the elaboration of the observable human phenomena which we are attempting to unify within a study of mind. These phenomena have been studied since the dawn of time. There are many monumental works of schlarship which unify the phenomena grouped into well-defined subfields. The only problem for AI workers surveying all these masterpieces is that none of the authors are committed to computational models. Indeed, they would no doubt laugh at anyone who suggested that their work could be reduced to a Turing Machine compatible notation. > This is not to say that AI research should halt But AI research could at least be disciplined to study the existing work on the phenomena they seek to study. Exploratory, anarchic, uninformed, self-indulgent research at public expense could be stopped. (and not just in AI, although I've never seen such a lack of discipline and scholarship anywhere else outside of popular history and futorology, neither of which attract public funds). > or that computers are not useful in studying human > intelligence. (They are indispensable.) Yes (no). They have proved useful in many areas of study. They have never been used at all in others, beacuse they have not been able to offer anything worthy of attention. > For one example of this new way of thinking, see the recent book by the > linguist George Lakoff, entitled "Women, Fire, and Dangerous Things." Does he use computers? >I believe the great success of AI has been in showing that >the old dualistic separation of mind and body is totally >inadequate to serve as a basis for an understanding of human intelligence. How can you attribute the end of dualism to AI research. This is a historical statement which should be backed up by references to specific pieces of work in AI. I doubt that anything emerging from AI (rather than the disciplines of Cognitive Science) -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ End of AIList Digest ******************** 27-Oct-87 21:58:09-PST,13838;000000000000 Mail-From: LAWS created at 27-Oct-87 21:49:37 Date: Tue 27 Oct 1987 21:43-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #250 - Cybernetics, Education, Neuromorphic Simulators To: AIList@SRI.COM AIList Digest Wednesday, 28 Oct 1987 Volume 5 : Issue 250 Today's Topics: Queries - NIL & Literature Classification & Parallel Logic Programming and Architectures, Definitions - Cybernetics, Education - Introductory Lisps and Prologs, Neuromorphic Systems - Simulator Sources, References - Chaos Theory ---------------------------------------------------------------------- Date: Tue, 27 Oct 87 06:54:35 -0500 From: johnson Subject: NIL (the lisp) where can i get a copy (of the source code for) NIL (the lisp implementation)? does anyone out there have a small (minimal) fast lisp in C with free or at least royalty-free source code ? thanks, johnson@UDEL ------------------------------ Date: 26 Oct 87 15:53:00 GMT From: mcvax!unido!uklirb!noekel@uunet.uu.net Subject: Literature classification - (nf) Hi everybody, we're currently building a AI bibliography and are still searching for a suitable classification/key word scheme. If there are any schemes that have gained wide-spread use in the AI community I would be very interested to learn about them. Obviously adopting such an existing scheme would be the sensible thing to do since in this case it would be much easier to merge our bibliography with others. Hints and pointers are welcome. If I get buckets of answers, I'll summarize to the net. Thanks in advance Klaus Noekel Universitaet Kaiserslautern Fachbereich Informatik Postfach 3049 6750 Kaiserslautern West Germany UUCP: ...!mcvax!unido!uklirb!noekel ------------------------------ Date: 27 Oct 87 14:49:00 EST From: Innes (I.A.) Ferguson Subject: Parallel logic programming and architectures I am currently in the process of trying to track down which schools are doing graduate research in the area of parallel logic programming and/or related machine architectures, and have come up with about a dozen so far. If anybody attends or knows of any graduate schools doing work in this area, I would very much like to hear from them. If I get enough response, I'll make up a list and post it on the net. Thanks in advance. Innes A. Ferguson, BNR Ltd., Ottawa, Canada tel.: (613) 727-2586 NETNORTH: iaf@bnr ------------------------------ Date: 26 Oct 87 09:56:26 GMT From: speedy!honavar@speedy.wisc.edu (A Buggy AI Program) Subject: Cybernetics, some definitions In article <3861@venera.isi.edu> smoliar@vaxa.isi.edu.UUCP (Stephen Smoliar) writes: >In article <8300006@osiris.cso.uiuc.edu> goldfain@osiris.cso.uiuc.edu writes: >> On the other hand, whatever became of the term "cybernetics" that Norbert >>Weiner coined long ago? I thought its definition was quite suitable for >>denoting this research. ... Some definitions (From "Cybernetic Medley" by Pekelis, MIR Publishers, Moscow, 1986 - which by the way, is an eminently readable book): "study of control and communication in machines and human beings" - -- Norbert Weiner, USA. "a science concerned with the study of systems of any nature which are capable of receiving, storing, and processing information so as to use it for control" -- Academician A. N. Kolmogorov, USSR. "the art of securing efficient operation" -- L. Couffignal, France. "a general theory of causality which is interpreted accurately to the part of isomorphism" -- A. Markov, Associate fellow, USSR Academy of Sciences. "a science concerned with the control of sophisticated dynamic systems, which is theoretically based on mathematics and logic, and practically, on the use of means of automation, electronic computers of primarily control and data-processing types" -- Axel Berg "a science concerned with the laws of receiving, storing, transmitting, and processing information in sophisticated systems of control" -- Academician V. Glushkov, USSR. "a science concerned with systems that have vitality, that is, which behave so as to survive" -- Stafford Beer, British mathematician -- Vasant Honavar (honavar@speedy.cs.wisc.edu) ------------------------------ Date: 26 Oct 87 06:02:22 GMT From: rocky!wagner@labrea.stanford.edu (Juergen Wagner) Subject: Re: Suggestions for Course In my opinion, Prolog and AI are not that much interwoven as they are, just because some people in a small room somewhere decided to use Prolog for their (so-called) AI problems, but because Prolog is SUITABLE and ADEQUATE for this class of problems for a number of reasons. One shouldn't argue that Prolog is no good to be taught in an AI class because of bad experience with this type of courses. If fact, requests for information on how to teach these courses will hopefully improve them. Juergen Wagner, (USENET) gandalf@portia.stanford.edu Center for the Study of Language and Information (CSLI), Stanford CA ------------------------------ Date: 26 Oct 87 20:44:38 GMT From: devvax!jplpro!des@elroy.jpl.nasa.gov (David Smyth) Subject: Re: Suggestions for Course In article <1746@unc.cs.unc.edu> bts@unc.UUCP (Bruce Smith) writes: >Turbo Prolog for an AI course? Why not FORTRAN, for that matter? >Quoting (without permission) from Alan Bundy's Catalog of AI Tools: > > 2. AI programs tend to be very poorly constructed, ... > ... FORTRAN provides a special mechanism > for achieving this, the so-called GOTO statement. > > 3. FORTRAN provides a very efficient data structure, the > array, which is particularly useful if, for example, one > wishes to process a collection of English sentences each > of which has the same length. Well, I must admit that I tried exactly this: I used FORTRASH for my homework assignments in my "Intro to AI" course. The professor's response: "What is this? Some kind of a joke?" I explained that SUCKTRAN with a stack is Turing equivalent, and so there was nothing he could do in Lisp that I could not do in SUCKTRASH. Besides, the Lisp system at school ran on one of those dinosaurs that all the undergrads had to use, so it was predictably unreliable and SLOW. I, on the otherhand, had acess to all these neat-o bitchen machines at work with various FORTRAN viruses, which both worked and had good response. It did take me awhile to develop libraries I needed, like variable length array support (Lisp, at some level, has to be concerned about running out of contiguous space too, you know - or didn't you %^) Anyway, I passed. Probably got an A too. ------------------------------ Date: 26 Oct 87 20:30:31 GMT From: decvax!necntc!ci-dandelion!bunny!mdf0@ucbvax.Berkeley.EDU (Mark Feblowitz) Subject: Re: Suggestions for Course With regard to the use of Turbo Prolog, I would like to make bring up the standard argument against "experience polution." The notion that an educational tool can afford to be non-standard or inferior because it is "only for beginners" assumes that the beginner will be able to reformulate his/her attitudes or habits when the appropriate time arrives. For this reason, I recommend the use of a Prolog implementation that is more compatible with C&M Prolog. My favorite Prolog for the PC is Arity Prolog. It is what I consider to be a production quality Prolog development environment. It: is FAST, links to assembly code or C has a full virtual memory system can run either compiled or interpreted ... The compiler comes with a "lint" facility for static precompilation analysis of typical Prolog programming errors. There is a low-end interpreter for the beginner, and I believe that there are educational discounts available (contact Arity Prolog to verify this). Although I have encountered a few bugs, the folks at Arity have been quite supportive in fixing these bugs. I am looking forward to their upcoming version 5, which aparently has an integrated editor and enhanced user-interface capabilities, among other things. Arity Prolog is located in Concord, MA, (617) 371-1243. I am NOT a representative of, nor do I have any financial interest in Arity Corp. I am merely a satisfied user of Arity Prolog. Mark Feblowitz GTE Laboratories, Inc., 40 Sylvan Rd. Waltham, MA 02254 (617) 466-2947 CSNET: feblowitz@GTE-LABS.CSNET UUCP: feblowitz@bunny.UUCP old UUCP: harvard!bunny!mdf0 -- Mark Feblowitz GTE Laboratories, Inc., 40 Sylvan Rd. Waltham, MA 02254 (617) 466-2947 CSNET: feblowitz@GTE-LABS.CSNET UUCP: feblowitz@bunny.UUCP old UUCP: harvard!bunny!mdf0 ------------------------------ Date: 27 Oct 87 09:28:32 est From: Walter Hamscher Subject: Suggestions for Course It seems to me that your list of reasons for using Lisp or Prolog in an AI class didn't include the single substantive reason for using Lisp, which is that that its use of just a couple of primitive data types to represent both data and code make it particularly easy to writing specialized languages and interpreters or compilers for them. Especially pattern-directed invocation languages. The reason for using Prolog is that it embeds one particular pattern-directed invocation paradigm for writing AI programs and makes that extremely fast, although clearly your experience with DCG notation suggests that it too has the unity of code and data that is helpful in constructing alternative interpreters. The reason it is important to be able to build specialized languages and interpreters is that those are what makes it possible to build specialized representations appropriate for different problems. And the engineering of appropriate representations is fundamental to AI. (which is not to claim that we know how to do it very well yet :-) ------------------------------ Date: 27 Oct 87 15:42:21 GMT From: gatech!hubcap!steve@bloom-beacon.mit.edu ("Steve" Stevenson) Subject: Re: Suggestions for Course in article <10475@duke.cs.duke.edu>, gleicher@duke.cs.duke.edu (Michael Gleicher) says: Xref: hubcap comp.lang.prolog:357 comp.ai:845 One of the reasons to use prolog is to give my students another language model. It also motivates the study of certain topics in resolution. The question of what's right or wrong with the exact prolog used is less important in my mind as long as the students see that TurboPascal is not the world. -- Steve (really "D. E.") Stevenson steve@hubcap.clemson.edu Department of Computer Science, (803)656-5880.mabell Clemson University, Clemson, SC 29634-1906 ------------------------------ Date: 27 Oct 87 22:17:49 GMT From: joglekar@riacs.edu (Umesh D. Joglekar) Reply-to: joglekar@hydra.riacs.edu (Umesh D. Joglekar) Subject: Re: Introductory books on Lisp Try .. ANATOMY OF LISP - By Allen ------------------------------ Date: Tue, 27 Oct 87 17:28:11 est From: ah4h+@andrew.cmu.edu (Andrew Hudson) Subject: Re: neuro sources This is in response to a query for connectionist simulator code. Within a month, one of the most comprehensive back propagation simulators will be available to the general public. Jay McClelland and David Rumelhart's third PDP publication, Exploring Parallel Distributed Processing: A Handbook of Models, Programs, and Exercises will be available from MIT Press. C source code for the complete backprop simulator, as well as others, is supplied on two MS-DOS format 5 1/4" floppy discs. The simulator, called BP, comes with the necessary files to run encoder, xor, and other problems. It supports multiple layer networks, constrained weight, and sender to receiver options. The handbook and source code can be ordered from MIT Press at the address below. The cost for both is less than $30. Why spend thousands more for second best? The MIT Press 55 Hayward Street Cambridge, MA 02142 Another version of the BP simulator which is not yet generally available to the public has been modified to take full advantage of the vector architecture of the Convex mini-supercomputer. For certain applications this gives speed increases of 30 times that of a VAX 11/780. A study is underway to see how well BP will perform on a CRAY XMP-48. - Andrew Hudson ah4h@andrew.cmu.edu.arpa Department of Psychology Carnegie Mellon 412-268-3139 Bias disclaimor: I work for Jay, I've seen the code. ------------------------------ Date: 27 Oct 87 15:45:21 GMT From: ssc-vax!dickey@beaver.cs.washington.edu (Frederick J Dickey) Subject: Re: The success of AI (misunderstandings) In article <8710260721.AA26918@ucbvax.Berkeley.EDU>, nick@MC.LCS.MIT.EDU writes: > For a good intro to chaos theory, see the article by Farmer, > Packard, et. al. in Scientific American December 86.. Recently, on popular book on chaos has been published. Its title is "Chaos" and the author is Gleick. Sorry, I don't remember any more details. It seems to be a good book, but I don't have any idea if professional chaoticians would like it. ------------------------------ End of AIList Digest ******************** 27-Oct-87 22:14:18-PST,14042;000000000000 Mail-From: LAWS created at 27-Oct-87 22:10:53 Date: Tue 27 Oct 1987 22:06-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #251 - Kolmogorov, Supercomputing, Methodology To: AIList@SRI.COM AIList Digest Wednesday, 28 Oct 1987 Volume 5 : Issue 251 Today's Topics: Obituary - A.N. Kolmogorov, Review - Spang Robinson Report on Supercomputing, V1 N2, Comments - AI Methodology ---------------------------------------------------------------------- Date: 26 Oct 1987 10:18:13-EST (Monday) From: Leonid Levin Reply-to: TheoryNet List Subject: The death of A.N. Kolmogorov. I just learned that in Moscow died Andrei Nikolayevich Kolmogorov - a great mathematician who also made crucial contributions to Theoretical Computer Science, Probability and Statistics, Information Theory and other fields. He also was one of those rare people whose personal integrity influenced ethical and human standards to the extent possible under the difficult conditions of a totalitarian state. Any telegrams from organizations and persons who appreciated the contributions of A.N. Kolmogorov will be gratefully received. They may be directed to Moscow University, The Academy of Sciences of the U.S.S.R. and the widow Anna Dmitriyevna Kolmogorov (117234, Moscow, Moscow University, korpus (building) L, apartment 10, USSR). ------------------------------ Date: Sun, 25 Oct 1987 12:23 CST From: Leff (Southern Methodist University) Subject: Review - Spang Robinson Report on Supercomputing, V1 N2 Summary of Spang Robinson Report on Supercomputing and Parallel Processing Volume 1 , No. 2 The lead article is on Hypercube based systems emphasizing the offerings from Intel, NCUBE and Floating Point Systems. The first implementation of the Hypercube based architectures was in the Soviet Union in the 1970's. Spang Robinson estimates revenues from 15 to 20 million dollars per year from Hypercube companies in 1987. OVUM predicts 1.15 billion dollar revenue from hypercubes in 1990. Intel Scientific Computers has benchmarked its 80286 based vector processor with 32 nodes against a Cray X-MP on a Fluid Dynamics Code and achieved equivalent performance. The iPSC/2 has 80386 and 80387 in each node with 512KB. They use Unix V.3 and a Concurrent Workbench which allows multiple users to share the node. System prices ranges from $165,000 to $1.76 million for systems from 16 nodes to 128 nodes. The 64 node vector processor sells for $929 thousand. Floating Point Systems reports 1.2 GigaFlps on their T200 system. They have made a total of ten installations. The node contains a T414 Transputer as a control node, a vector processor, 1 meg of RAM and four communication links. Each node takes one printed circuit board and eight nodes are grouped with an 80 meg disk storage unit. A T-20 costing $400,000 contains 16 nodes and is rated at 192 MFLOPS. A T200 with 128 nodes costs 3 millions. They use a DEC micro VAX II to host the system and some system programming is done in OCCAM. One is installed at Clemson University where they are install SPICE. Ametek sold several S-14 systems and is working on a second-geeneration product with announcements planned before the end of 1987. They have doubled their building space. NCUBE uses a proprietary chip at each node with 512 kilobytes of memory and twenty-two DMA channels. 64 nodes are on a single printed circuit board. The NCUBE 7 goes to 128 nodes ($350,000 with 500 MB DISK) and the NCUBE 10 goes to 1024 nodes ($1.7 million). An NCUBE 4 that fits in a PC is from $20,000 to $60,000. They use a UNIX like operating system. A total of 70 installations have been made with half in Universities. ___________________________________________________________________ Book Review: The Supercomputer Era by Sidney Karin and Norris Parker Smith _________________________________________________________________________ Discussions of Cray Research changes included the departure of Steve Chen. Steve Chen has formally announced a new corporation to continue the research. __________________________________________________________________________ The NSA has set up its own supercomputer development project, deciding that industry will not produce products meeting its need. ------------------------------ Date: 23 Oct 87 16:35:26 GMT From: umix!umich!dwt@uunet.UU.NET (Dave West) Reply-to: umix!zippy.eecs.umich.edu!dwt@uunet.UU.NET (David West) Subject: Re: Lenat's AM program In article <8710211650.AA18715@orstcs.CS.ORST.EDU> tgd@ORSTCS.CS.ORST.EDU (Tom Dietterich) writes: >The exact reasons for the success of AM (and for its eventual failure >to continue making new discoveries) have not been established. [...] > >The problem with all of these explanations is that they have not been >subjected to rigorous experimental and analytical tests, so at the >present time, we still (more than ten years after AM) do not >understand why AM worked! Some possible contributing reasons for this sort of difficulty in AI: 1) The practitioners of AI routinely lack access at the nuts-and-bolts level to the products of others' work. (At a talk he gave here three years ago, Lenat said that he was preparing a distribution version of AM. Has anyone heard whether it is available? I haven't.) Perhaps widespread availability and use of Common Lisp will change this. Perhaps not. 2) The supporting institutions (and most practitioners) have little patience for anything as unexciting and 'unproductive' as slow, painstaking post-mortems. 3) We still have no fruitful paradigm for intelligence and discovery. 4) We are still, for the most part, too insecure to discuss difficulties and failures in ways that enable others as well as ourselves to learn from them. (See an article on the front page of the NYTimes book review two or three weeks ago for a review of a book claiming that twentieth- century science writing in general is fundamentally misleading in this respect.) David West dwt@zippy.eecs.umich.edu ------------------------------ Date: 26 Oct 87 19:57:47 GMT From: ritcv!cci632!mdl@cs.rochester.edu (Michael Liss) Subject: Re: Goal of AI: where are we going? (the right way?) In article <285@usl> khl@usl.usl.edu.UUCP (Calvin K. H. Leung) writes: >I agree with the idea that there must be some mechanisms that our >minds are using. But the different reasoning methods (proba- >bilistic reasoning, for instance) that we are studying in the >area of AI are not the way one reasons: we never use the Bayes' >Theorem in our thinking process. The use of those reasoning >methods, in my point of view, will never help increase our under- >standing of human behavior. Because our minds just don't work >that way. I read an interesting article recently which had the title: "If AI = The Human Brain, Cars Should Have Legs" The author's premise was that most of our other machines that mimic human abilites do not do so through strict copying of our physical processes. What we have done, in the case of the automobile, is to make use of wheels and axles and the internal combustion engine to produce a transportation device which owes nothing tothe study of human legs. In the case of AI, he state that artificial intelligence should not be assumed to be the equivalent of human intelligence and thus, the disection of the human mind's functionality will not necessarily yield a solution to AI. He closes with the following: "And I suspect it [AI] will develop without reference to natural intelligence and should so develop. And I am sure it will not replace human thinking any more than the autombile replaces human walking." "Why am I so soft in the middle when the rest of my life is so hard?" -- P.Simon Mike Liss {rochester, ritcv}!cci632!mdl (716) 482-5000 ------------------------------ Date: 26 Oct 87 17:03:26 GMT From: net1!todd@sdcsvax.ucsd.edu (Todd Goodman) Subject: Re: The Success of AI In article <131@glenlivet.hci.hw.ac.uk> gilbert@hci.hw.ac.uk (Gilbert Cockton) writes: >"Better" concepts related to mind than those found in cog. sci. >already exist. The starting point is the elaboration of the observable human >phenomena which we are attempting to unify within a study of mind. These >phenomena have been studied since the dawn of time. There are many >monumental works of schlarship which unify the phenomena grouped into >well-defined subfields. The only problem for AI workers surveying all >these masterpieces is that none of the authors are committed to >computational models. Indeed, they would no doubt laugh at anyone who >suggested that their work could be reduced to a Turing Machine compatible >notation. Please, please, please give us a bibliography of these works. In fact a short summary would be great, along with the reasons that you find them to be better than any current models. Also if you could point out which are at odds with each and which you feel are "better" than others, then I would be greatly appreciative. This isn't a flame about your response to the earlier posting. I just want to take a look at the monumental works you're talking about. Todd Goodman todd@net1.ucsd.edu ...!{ucbvax|ihnp4}!sdcsvax!net1!todd ------------------------------ Date: 26 Oct 87 03:31:00 GMT From: uxc.cso.uiuc.edu!osiris.cso.uiuc.edu!goldfain@a.cs.uiuc.edu Subject: Re: The Success of AI > tsmith@gryphon.CTS.COM writes > Now here's the interesting point. If you were to come to me and say-- > "Smith, you have a year to develop an automaton that will play some > kind of major sport at a championship level, competing against humans. > Money is no object, and you can have access to all the world's > experts in AI and robotics, but you must design a robot that plays > championship X in a year's time. What is X?" I would say, without a > moment's hesistation, "tennis". > > Why? Of all the sports, tennis is the most bounded. It is played within > a very restricted area (unlike golf or even baseball), it is a > one-against-one sport (unlike football or soccer), the playing surfaces > (aside from Wimbledon) are the truest of all the major sports, and it > is indubitably the most boring of all the sports to watch (if not to > play). A perfect candidate for automation. > ---------------- Hmmm, by your own criterion, I would prefer table tennis, or to make life really easy, bowling. I had heard that a table-tennis playing robot has been developed that is really quite good. Bowling is really way too simple. (If what I have heard is correct, othello would also be a good choice - computers have already been claimed by some to outperform humans here, but it's not a major sport.) ------------------------------ Date: 27 Oct 87 02:06:56 GMT From: topaz.rutgers.edu!josh@rutgers.edu (J Storrs Hall) Subject: Re: The Success of AI > tsmith@gryphon.CTS.COM writes > Now here's the interesting point. If you were to come to me and say-- > "Smith, you have a year to develop an automaton that will play some > kind of major sport at a championship level, competing against humans. > Money is no object, and you can have access to all the world's > experts in AI and robotics, but you must design a robot that plays > championship X in a year's time. What is X?" I would say, without a > moment's hesistation, "tennis". Goldfain says bowling, which is a very good choice, being in a completely artificial environment. It might have (with ping-pong) the problem of not "really being a sport". If we define "major sport" as something done outside in real time against competition and often televised on major networks, I would have to go with the 50 yard dash. If we allow any olympic event, offhand sharpshooting looks promising, javelin throwing looks easy, shot put looks trivial. In fact, the more I think about it, tennis is probably one of the *hardest* sports to implement. I imagine a team of football-playing robots: they look something like tanks... The point in all this is obviously that in the history of replacing human effort with mechanical effort, brute force was the first success story. * * * * "The Yankees pitcher steps to the mound. It is a Cincinnati Milacron G97A22013 just brought up from the minors. Here's the pitch! Holy cow! A 957 mph fastball on the inside corner for strike one! ..." --JoSH ------------------------------ Date: 26 Oct 87 03:38:41 GMT From: imagen!atari!portal!cup.portal.com!tony_mak_makonnen@ucbvax.Berk eley.EDU Subject: Re: The success of AI (misunderstandings) this is exemplary of what happens when many perspectives enter the picture and words flow . I submit the following : It was Von Neuman himself ( I believe) who said that anything that can be calculated precisely i.e. mathematically can be done better by a computer . ( I think this should pass even by the most rabid hater of computers ) I note that man who is getting lambasted used the words computed and computational. I should think he would agree that if one began to talk of reflection , intuition and so on , the conversation would be totally different . Else are we to think that with great enough and intensive computation the machine will eventually exhibit awareness of itself as something that is .?! ------------------------------ End of AIList Digest ******************** 29-Oct-87 01:58:32-PST,11428;000000000000 Mail-From: LAWS created at 29-Oct-87 01:38:56 Date: Thu 29 Oct 1987 01:35-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #252 - Neural Network Review, UK Mail, Seminars To: AIList@SRI.COM AIList Digest Thursday, 29 Oct 1987 Volume 5 : Issue 252 Today's Topics: Journal - Neural Network Review, Binding - Transatlantic Netmail to UK, Seminars - Dependency-Directed Prolog (BBN) & Speech Recognition Using Connectionist Networks (UNISYS), Conference - Expert Systems in Business and Finance ---------------------------------------------------------------------- Date: Wed, 28 Oct 87 15:30:08 EST From: csed-1!will@hc.dspo.gov (Craig Will) Subject: Announcing Neural Network Review Announcing a new publication NEURAL NETWORK REVIEW The critical review journal for the neural network community Neural Network Review is intended to provide a forum for critical analysis and commentary on topics involving neural network research, applications, and the emerging industry. A major focus of the Review will be publishing critical reviews of the neural network literature, including books, individual papers, and, in New York Review of Books style, groups of related papers. The Review will also publish general news about events in the neural network community, including conferences, funding trends, and announcements of new books, papers, courses, and other media, and new hardware and software pro- ducts. The charter issue, dated October, 1987, has just been published, and contains a review and analysis of 11 articles on neural networks published in the popular press, a report on the San Diego conference, a report on new funding initia- tives, and a variety of other information, a total of 24 pages in length. The next issue, due in January, 1988, will begin detailed reviews of the technical literature. Neural Network Review is aimed at a national audience, and will be published quarterly. It is published by the Washington Neural Network Society, a nonprofit organization based in the Washington, D.C. area. Subscriptions to Neural Network Review are $ 10.00 for 4 issues, or $ 2.50 for a single copy. International rates are slightly higher. Rates for full-time students are $5.00 for 4 issues. (Checks should be payable to the Washington Neural Network Society). Subscription orders and inquiries for information should be sent to: Neural Network Review P. O. Box 427 Dunn Loring, VA 22027 For more information on Neural Network Review, send your physical, U. S. Postal mail address in a message to will@hc.dspo.gov (Craig Will). ------------------------------ Date: 28 Oct 87 16:46 PST From: hayes.pa@Xerox.COM Subject: transatlantic netmail to UK I recently had some correspondence about this with an informed UK source, and here is his statement about what is going on and why, and what the future should hold. Looks good. Pat Hayes ---------- The costs of transatlantic traffic, in both directions, through the UCL Arpanet gateway are borne by a UK funding agency, the Alvey Directorate/SERC . Darpa does not pay for messages originating in the USA and sent to the UK gateway, and UCL ( University College, part of London University ) has no way of charging individual American originators of messages. Some time ago, UCL needed to get more accurate statistics about UK usage to strengthen its case for more money to run the transatlantic link. To show that the gateway was a vital facility, UCL instigated the policy of requiring UK users to be properly authorised, ie officially registered as users. The cost of this bi-directional transatlantic traffic now exceeds the budget granted by Alvey/SERC to UCL. Appeals by UCL to SERC brought to light that much net traffic originating in the UK was being channelled through the very few `official' accounts. Moreover, UCL had no data on the number of US customers it serves. More recently, the increased cost of running the link has meant that UCL now wishes to track traffic originating in the USA, to help show the importance of the link. As USA to UK messages are not funded by any USA agency, then either the SERC pays for it all, via the UCL budget, or some form of charge-back to UK recipients must be instigated. This is the origin of the recent change in operating policy requiring USA users to be registered as collaborating with some specific UK group ie charging centre. Such a charge would then be allowable against individual SERC grants, rather than UCL picking up the total cost. There is no suggestion that any USA user will be refused authorisation. It is clear to all parties that this is not a satisfactory mechanism, either now or for the future. I am pleased to tell you that negotiations are now well advanced for a more permanent and sensible solution. The proposal is that the UK's SERC and the USA's NSF (more natural counterparts than Darpa) will instigate a new USA-UK link, properly jointly organised and funded for the benefit of academics. This link will, on the USA side, gateway the UK's Janet (the official name of the UK academic net) into most of the USA nets (arpa, NSF's own, Usenet etc. ) The present Arpanet-Janet link will continue until this improved NSF-Janet comes into service. There is no firm date for this yet, but I think that if people in the USA cooperate with UCL in the short term, and have a little patience and sympathy for UCL's predicament, then we should all be able to keep communicating via UCL until the next generation gateway comes into service. ------------------------------ Date: Tue 27 Oct 87 10:38:24-EST From: Marc Vilain Subject: Seminar - Dependency-Directed Prolog (BBN) BBN Science Development Program AI Seminar Series Lecture DEPENDENCY DIRECTED PROLOG Jeffrey Mark Siskind MIT Laboratory for Computer Science (also: summer intern at Xerox PARC) (Qobi@ZERMATT.LCS.MIT.EDU) BBN Labs 10 Moulton Street 2nd floor large conference room 10:30 am, Tuesday November 3 In this talk I will describe an implementation of pure Prolog which uses dependency directed backtracking as a control strategy for pruning the search space. The implementation uses a strategy whereby the Prolog program is compiled into a finite set of templates which characterize a potentially infinite boolean expression which is satisfiable iff there is a proof of the goal query. These templates are incrementally unraveled into a sequence of propositional CNF SAT problems and represented in a TMS which is used to find solutions using dependency directed backtracking. The technique can be extended to use ATMS-like strategies for searching for multiple solutions simultaneously. Two different strategies have been implemented for dealing with unification. The first compiles the unification constraints into SAT clauses and integrates them in the TMS along with the and/or goal tree produced by unraveling the templates. The second uses a separate module for doing unification at run time. This unifier is novel in that it records dependencies and allows nonchronological retraction. The interface protocol between the TMS and the unifier module has been generalized to allow integration of other "domains" of predicates, such as linear arithmetic and simple linear inequalities, to be built into the system while still preserving the soundness and completeness of the pure logical interpretation of Prolog. In the talk, time permitting, I will discuss the search prunning advantages of this approach and its relation to previous approaches, the implementation mechanism, and some recent work indicating the potential applicability of this approach to parsing with disjunctive feature structures, such as done with the LFG and related grammar formalisms. ------------------------------ Date: Tue, 27 Oct 87 15:35:57 EST From: finin@bigburd.PRC.Unisys.COM (Tim Finin) Subject: Seminar - Speech Recognition Using Connectionist Networks (UNISYS) AI Seminar UNISYS Knowledge Systems Paoli Research Center Paoli PA SPEECH RECOGNITION USING CONNECTIONIST NETWORKS Raymond Watrous Siemens Corporate Research and University of Pennsylvania The thesis of this research is that connectionist networks are adequate models for the problem of acoustic phonetic speech recognition by computer. Adequacy is defined as suitably high recognition performance on a representative set of speech recognition problems. Six acoustic phonetic problems are selected and discussed in relation to a physiological theory of phonetics. It is argued that the selected tasks are sufficiently representative and difficult to constitute a reasonable test of adequacy. A connectionist network is a fine-grained parallel distributed processing configuration, in which simple processing elements are interconnected by simple links. A connectionist network model for speech recognition has been defined called the TEMPORAL FLOW MODEL. The model incorporates link propagation delay and internal feedback to express temporal relationships. It has been shown that temporal flow models can be 'trained' to perform successfully some speech recognition tasks. A method of 'learning' using techniques of numerical nonlinear optimization has been demonstrated for the minimal pair "no/go", and voiced stop consonant discrimination in the context of various vowels. Methods for extending these results to new problems are discussed. 10:00am Wednesday, November 4, 1987 Cafeteria Conference Room Unisys Paloi Research Center Route 252 and Central Ave. Paoli PA 19311 -- non-UNISYS visitors who are interested in attending should -- -- send email to finin@prc.unisys.com or call 215-648-7446 -- ------------------------------ Date: Thu 29 Oct 87 01:08:12-PST From: Ken Laws Reply-to: AIList-Request@SRI.COM Subject: Conference - Expert Systems in Business and Finance John Feinstein [(703) 934-3280] asked me to send out a notice about the first annual conference on expert systems in business and finance -- but I see that John Akbari submitted a description in AIList V5 N246, Oct. 25. I'll just repeat that it's at the Penta Hotel in New York City, November 10-12, 1987, $525. Call (609) 654-6266. ------------------------------ End of AIList Digest ******************** 29-Oct-87 21:51:58-PST,9003;000000000000 Mail-From: LAWS created at 29-Oct-87 21:43:32 Date: Thu 29 Oct 1987 21:41-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #253 - LISP, NIL, Msc. To: AIList@SRI.COM AIList Digest Friday, 30 Oct 1987 Volume 5 : Issue 253 Today's Topics: Queries - Symbolic and Algebraic Computation Text & Prediction-Producing Algorithms & Mental Models, Bibliographies - Literature Classification, AI Tools - LISP on the AMIGA & NIL (the LISP) & Character Recognition, Correction - Spang Robinson Report on Supercomputing, V1 N2 ---------------------------------------------------------------------- Date: 26 Oct 1987 13:54:48 EST From: Walter.Daugherity@LSR Subject: Symbolic and Algebraic Computation Text There are a number of interesting AI people and projects here at Texas A & M and we are expanding. [... looking for a CS department head.] Also, I will be teaching a graduate course in Symbolic and Algebraic Computation and am looking for useful textbooks, proceedings, citations, etc., if you know of any. Thanks, Walter BITNET: WCD7007@TAMLSR WCD7007@TAMSIGMA CSNET: WCD7007%LSR%TAMU@CSNET-RELAY WCD7007%SIGMA%TAMU@CSNET-RELAY Paper mail: Dr. Walter C. Daugherity Texas A & M University Department of Computer Science Zachry Engineering Center, Room 238 College Station, Texas 77843-3112 U.S.A. _________ ------------------------------ Date: 29 Oct 87 04:22:27 GMT From: mind!eliot@princeton.edu (Eliot Handleman) Subject: Prediction-producing Algorithms I am looking for any work done on predictive algorithms - by which I mean something that, given some input, is able to make a reasonable stab at a plausible continuation. I am decidedly not interested in things which compute transition probablities. Something which is able to generated some pattern of inference is more up my alley. For example, if I fed the pattern a a b a a b a into this thing, I would expect to get back a b as the most reasonable thing to expect. Any pointers to articles, dissertations, texts, programs etc would be extremely helpful. Please ship your replies to me directly, and many thanks in advance. ------------------------------ Date: 24 Oct 87 08:39:37 GMT From: cunyvm!byuvax!fordjm%psuvm.bitnet@ucbvax.Berkeley.EDU Subject: Mental Models I am getting ready to conduct a literature search to learn more about mental models from a cognitive psychology perspective. I am familiar with Gentner & Stevens' (1983) "Mental Models" and P. Johnson-Laird's (1983) book by the same name. Can anyone point me to an existing bibliography or recent references on this topic since 1983? Also, is anyone out there currently doing research in this area? Although I would find general references useful, I am particularly interested in applications of mental models theories in instructional/educational psychology and measurement. Please send responses to me via e-mail and I will summarize to the net. Thanks in advance. John M. Ford fordjm@byuvax.bitnet ------------------------------ Date: 28 Oct 87 15:32:14 GMT From: sunybcs!rapaport@ames.arpa (William J. Rapaport) Subject: Re: Literature classification - (nf) In article <23600004@uklirb.UUCP> noekel@uklirb.UUCP writes: > >we're currently building a AI bibliography and are still searching for a >suitable classification/key word scheme. If there are any schemes that have >gained wide-spread use in the AI community I would be very interested to >learn about them. I don't know of any offhand, but here's an idea for getting one started: why not use (or suitably modify) the list of entries in the new Encyclopedia of Artificial Intelligence (ed. S. C. Shapiro; John Wiley & Sons, 1987)? William J. Rapaport Assistant Professor Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3181 uucp: ..!{ames,boulder,decvax,rutgers}!sunybcs!rapaport internet: rapaport@cs.buffalo.edu [if that fails, try: rapaport%cs.buffalo.edu@relay.cs.net or: rapaport@buffalo.csnet ] bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: 28 Oct 87 19:20:52 GMT From: super.upenn.edu!eecae!nancy!umix!tardis!ronin!mike@rutgers.edu (Michael Nowak) Subject: Re: LISP on the AMIGA. In article <2561@cbmvax.UUCP> phillip@cbmvax.UUCP (Phillip Lindsay GUEST) writes: >I would like to hear from people working on anything related to LISP and/or >AI on the Amiga. This is important since I am trying to solicit a port of >a LISP product. Any general interest also welcome. (the more bullets the better) I bought the MCC Lisp awhile back to use in my AI class and it was generally useful for that. What would be really nice is an implementation of Common Lisp for the Amiga. Michael Nowak ------------------------------ Date: 28 Oct 87 19:01:45 GMT From: wagner@rocky.STANFORD.EDU (Juergen Wagner) Reply-to: gandalf@portia.stanford.edu (Juergen Wagner) Subject: Re: NIL (the lisp) >where can i get a copy (of the source code for) NIL (the lisp implementation)? Contact MIT AI Lab, Glenn Burke. That's where I got a copy of NIL from (about three years ago). But... >does anyone out there have a small (minimal) fast lisp in C with >free or at least royalty-free source code ? ...NIL is neither small, nor minimal. At least the version I worked with used to eat up a fair amount of CPU time (especially when I compiled LISP code). I don't know if there is a version of NIL under UNIX, I only know one under VMS. If you are looking for a CommonLISP system which is reasonably small (minimal), which provides the standard language capabilities (CLtL) plus some extensions, which allows for dynamic loading of C modules (and thereby e.g. interfacing to window systems), and which is royalty-free (non-commercial use), I suggest Kyoto CommonLISP (KCL). Read comp.lang.lisp for more details on how to get a copy of KCL (via anonymous FTP, direct order, etc.). Juergen Wagner, (USENET) gandalf@portia.stanford.edu Center for the Study of Language and Information (CSLI), Stanford CA ------------------------------ Date: 28 Oct 87 15:38:42 GMT From: ihnp4!alberta!auvax!kevinc@ucbvax.Berkeley.EDU (Kevin Barry Crocker) Subject: Re: Character recognition In article <2984@phri.UUCP>, roy@phri.UUCP (Roy Smith) writes: > In article <641@zen.UUCP> vic@zen.UUCP (Victor Gavin) writes: > > I have been asked to write some software which can (given an image > > produced by the scanner) reproduce the original text of the paper in a > > machine readable form. > > I don't know much about it, but a company called DEST markets a > 300-dpi scanner for the Macintosh (and, I think, IBM-PC) for about $2k, This may not be relevant to all, but a recent issue of PC Magazine does a review of both Desktop Publishing and Scanners for the PC Market. The issue is Volume 6 Number 17 October 13, 1987. Now, I realize that for Mac users this may not be totally relevant but some of these companies may make suitable software to make thier product usable on the Mac - especially those that link to PageMaker. In fact I seem to remember some vendors products being touted as both market products. ihnp4!alberta!auvax!kevinc (Kevin Crocker Athabasca University) Do our employers have opinions or is that what we get paid for! ------------------------------ Date: 29 Oct 87 15:27:04 GMT From: steve@hubcap.clemson.edu ("Steve" Stevenson) Subject: Correction to Review - Spang Robinson Report on Supercomputing, V1 N2 in article <8710280643.AA22004@ucbvax.Berkeley.EDU>, E1AR0002@SMUVM1.BITNET (Leff, Southern Methodist University) says: > > Summary of Spang Robinson Report on Supercomputing and Parallel Processing > Volume 1 , No. 2 > > Floating Point Systems reports 1.2 GigaFlps on their T200 system. > .... One is installed at Clemson University where they are installing] > SPICE. Actually, we are writing a special FORTRAN and C for Hypercubes which is also being used for IMPLEMENTING SPICE. A new sparse solver is being developed by Dan Warner in MathSci. Roy Pargas and Keith Allen are also involved in mapping algorithms. I'm heading the language work. -- Steve (really "D. E.") Stevenson steve@hubcap.clemson.edu Department of Computer Science, (803)656-5880.mabell Clemson University, Clemson, SC 29634-1906 ------------------------------ End of AIList Digest ******************** 29-Oct-87 22:14:45-PST,13499;000000000000 Mail-From: LAWS created at 29-Oct-87 21:53:40 Date: Thu 29 Oct 1987 21:51-PST From: AIList Moderator Kenneth Laws Reply-to: AIList@SRI.COM US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList V5 #254 - AI Methodology To: AIList@SRI.COM AIList Digest Friday, 30 Oct 1987 Volume 5 : Issue 254 Today's Topics: Comments - Methodology & The Success of AI ---------------------------------------------------------------------- Date: 26 Oct 87 19:26:01 GMT From: rosevax!rose3!starfire!merlyn@uunet.uu.net (Brian Westley) Subject: Re: The Success of AI In one article... > But AI research could at least be disciplined to study the existing work > on the phenomena they seek to study. Exploratory, anarchic, > uninformed, self-indulgent research at public expense could be stopped. and, in another article... >..Thus, I am not avoiding hard work; I am avoiding >*fruitless* work... > -- > Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN Tell me, how do you know WHICH AI methods WILL BE fruitless? You certainly must know, for you to call it anarchic, uninformed, and self-indulgent (but why 'exploratory' is used as a put-down, I'll never know - I guess Gilbert already knows how to build thinking machines, and just won't tell us). Research is like advertising - most of the money spent is fruitless, but you won't KNOW that until after you've TRIED it. (Of course it isn't entirely wasted; you now know what doesn't work). Fortunately, you have not convince me nor many other people that your view is to be held paramount, and all other avenues of work are doomed to failure. By the way, I am not interested in duplicating or otherwise developing models of how humans think; I am interested in building machines that think. You may as well tell a submarine designer how difficult it is to build artificial gills - it's irrelevant. --- Merlyn LeRoy "Anything a computer can do is immediately removed from those activities that require thinking, such as calculations, chess, and medical diagnoses." ------------------------------ Date: 28 Oct 87 15:06:21 GMT From: ig!uwmcsd1!uwm-cs!litow@jade.Berkeley.EDU (Dr. B. Litow) Subject: AI Recently postings have focused on the topic: 'AI - success or failure'. Some postings have been concerned with epistemological or metaphysical matters. Other postings have taken the view that AI is a vast collection of design problems for which much of the metaphysical worry is irrelevant. Based upon its history and current state it seems to me that AI is an area of applied computer science largely aimed at design problems. I think that AI is an unfortunate moniker because AI work is basically fuzzy programming (more accurately the design of systems supporting fuzzier and fuzzier programming) where the term 'fuzzy' is not being used in a pejorative sense. All of the automation issues in AI work are support issues for really fuzzy programming i.e. where humans can extend the interface with automata so that human/automata interaction becomes increasingly complex and 'undisciplined'. Thus in a large sense AI is the frontier part of software science. It could be claimed that at some stage of extension the interface becomes so complex (by human standards at the time) that cognition can be ascribed to the systems. Personally I doubt this will happen. On the other hand the free use of play-like interfaces must have unforeseeable and gigantic consequences for humans. This is where I see the importance of AI. I distinguish between cognitive studies and AI. The metaphysics belongs to the former,not the latter. ------------------------------ Date: 28 Oct 87 18:04:45 GMT From: tgd@orstcs.cs.orst.edu (Tom Dietterich) Subject: Re: Lenat's AM program David West (dwt@zippy.eecs.umich.edu) writes: Some possible contributing reasons for this sort of difficulty in AI: 1) The practitioners of AI routinely lack access at the nuts-and-bolts level to the products of others' work. (At a talk he gave here three years ago, Lenat said that he was preparing a distribution version of AM. Has anyone heard whether it is available? I haven't.) Perhaps widespread availability and use of Common Lisp will change this. Perhaps not. In the biological sciences, publication of an article reporting a new clone obligates the author to provide that clone to other researchers for non-commercial purposes. I think we need a similar policy in computer science. Publication of a description of a system should obligate the author to provide listings of the system (a running system is probably too much to ask for) to other researchers on a non-disclosure basis. 2) The supporting institutions (and most practitioners) have little patience for anything as unexciting and 'unproductive' as slow, painstaking post-mortems. 3) We still have no fruitful paradigm for intelligence and discovery. 4) We are still, for the most part, too insecure to discuss difficulties and failures in ways that enable others as well as ourselves to learn from them. (See an article on the front page of the NYTimes book review two or three weeks ago for a review of a book claiming that twentieth- century science writing in general is fundamentally misleading in this respect.) I disagree with these other points. I think the cause of the problem is lack of methodological training for AI and CS researchers. Anyone could have reimplemented an approximation of AM based on the published papers anytime in the past decade. I think the fact that people are now beginning to do this is a sign that AI is becoming methodologically healthier. A good example is the paper Planning for Conjunctive Goals by D. Chapman in Artificial Intelligence, Vol 32, No. 3, which provides a critical review and rational reconstruction of the NOAH planning system. I encourage all students who are looking for dissertation projects to consider doing work of this kind. --Tom ------------------------------ Date: Thu 29 Oct 87 00:25:55-PST From: Ken Laws Subject: Gilding the Lemon Tom Dietterich suggests that AI students should consider doing critical reviews and rational reconstructions of previous AI systems. [There, isn't a paraphrase better than a lengthy quotation?] I wouldn't discourage such activities for those who relish them, but I disagree that this is the best way for AI to proceed AT THE PRESENT TIME. Rigorous critical analysis is necessary in a mature field where deep understanding is needed to avoid the false paths explored by previous researchers. I don't claim that shallow understanding is preferable in AI, but I do claim that it is adequate. AI should not be compared to current Biology or Psychology, but to the heyday of mechanical invention epitomized by Edison. We do need the cognitive scientists and logicians, but progress in AI is driven by the hackers and the graduate students who "don't know any better" than to attempt the unreasonable. Progress also comes from applications -- very seldom from theory. The "neats" have been worrying for years (centuries?) about temporal logics, but there has been more payoff from GPSS and SIMSCRIPT (and SPICE and other simulation systems) than from all the debates over consistent point and interval representations. The applied systems are ultimately limited by their ontologies, but they are useful up to a point. A distant point. Most Ph.D. projects have the same flavor. A student studies the latest AI proceedings to get a nifty idea, tries to solve all the world's problems from his new viewpoint, and ultimately runs into limitations. He publishes the interesting behaviors he was able to generate and then goes on the lecture circuit looking for his next employment. The published thesis illuminates a new corner of mankind's search space, provided that the thesis advisor properly steered the student away from previously explored territory. An advisor who advocates duplicating prior work is cutting his students' chances of fame and fortune from the discovery of the one true path. It is always true that the published works can be improved upon, but the original developer has already gotten 80% of the benefit with 20% of the work. Why should the student butt his head against the same problems that stopped the original work (be they theoretical or practical problems) when he could attach his name to an entirely new approach? I am not suggesting that "artificial intelligence" will ever be achieved through one graduate student project or by any amount of hacking. We do need scientific rigor. I am suggesting that we must build hand-crank phonographs before inventing information theory and we must study the properties of atoms before debating quarks and strings. Only when we have exploited or reached impass on all of the promising approaches will there be a high probability that critical review of already explored research will advance the field faster than will trying something new. [Disclaimer: The views expressed herein do not apply to my own field of computer vision, where I'm highly suspicious of any youngster trying to solve all our problems by ignoring the accumulated knowledge of the last twenty years. My own tendency is toward critical review and selective integration of existing techniques. But then, I'm not looking for a hot new Ph.D. topic.] -- Ken Laws ------------------------------ Date: Wed, 28 Oct 87 08:36:34 -0200 From: Eitan Shterenbaum Subject: Success of AI Had it ever come into you mind that simulating/emulating the human brain is NP problem ? ( Why ? Think !!! ). Unless some smartass comes out with a proof for NP=P yar can forget de whole damn thing ... Eitan Shterenbaum (* As far as I know one can't solve NP problems even with a super-duper hardware, so building such machine is pointless (Unless we are living on such machine ...) ! *) Eitan ------------------------------ Date: 29 Oct 87 13:15:51 GMT From: Gilbert Cockton Reply-to: Gilbert Cockton Subject: Re: THE MIND In article <8710120559.AA17517@ucbvax.Berkeley.EDU> UUCJEFF@ECNCDC.BITNET writes: >I read some of the MIND theories espoused in the Oct 2 list, and am >frankly disappointed. All those debates are based on the Science vs >Mysticism debates that were going on 10 years ago when I was an undergrad. Isn't it a shame that so many people in AI are so ignorant of the substance of these debates? >5) AI should concern itself with solving problems, discovering new ways to >solve and conceptialize problems. It is not as glamorous as making >artificial souls, but more practical and fruitful. Fortunately, this highly sensible view is attracting more support, and, with luck, it should establish itself as the raison d'etre of AI research. A change of name would help (viz demise of cyberbetics), despite the view of many old hands (e.g. Simon), that they wouldn't have chosen the name, but we are stuck with it now. I can't see how any sensible person would want to stick with a term with such distasteful connotations. However, this orientation for post-AI advanced computer applications research needs extension. It is not enough to develop new computerised support for new problem solving techniques. Research is also needed into the comprehensibility, ease of learning and validity of these techniques. Determinants of their acceptability in real organisational settings are also a vital research topic. Is research in medical expert systems, for example, worth public funding when it seems that NO medical expert system is being used in a real clinical setting? What sorts of systems would be acceptable? Similarly, the theorem prover based proof editors under development for software engineering seem to require knowledge and skills which few practising software professionals will have time to develop, so one can't really see proof editors developing into real work tools until a major shift in their underlying models occur. Such a user-oriented change of direction is a major problem for AI researchers, as few of them seem to have any real experience of succesfully implementing a working system and installing it in a real organisational setting, and then maintaining it. DEC's XCON is one of the few examples. How much is PROSPECTOR used these days? -- Gilbert Cockton, Scottish HCI Centre, Ben Line Building, Edinburgh, EH1 1TN JANET: gilbert@uk.ac.hw.hci ARPA: gilbert%hci.hw.ac.uk@cs.ucl.ac.uk UUCP: ..{backbone}!mcvax!ukc!hwcs!hci!gilbert ------------------------------ Date: 29 Oct 87 00:23:00 GMT From: uxc.cso.uiuc.edu!osiris.cso.uiuc.edu!goldfain@a.cs.uiuc.edu Subject: Re: The Success of AI (continued, a Who says that ping-pong, or table tennis isn't a sport? Ever been to China? ------------------------------ End of AIList Digest ********************