neuron-request@HPLMS2.HPL.HP.COM ("Neuron-Digest Moderator Peter Marvit") (12/14/90)
Neuron Digest Thursday, 13 Dec 1990 Volume 6 : Issue 71 Today's Topics: tech rep available on evolving neural networks learning a synaptic learning rule. TR available. tech. rep. available on evolving trail-following organisms Dissertation abstract & Technical report Preprint about scaled conjugate gradient available. graduate study in neural networks TR on Children's Acquisition of Irregular Past Tense available IJCNN-91-Seattle paper deadline is coming up! reprints available left-out info: IJCNN-91-Seattle FTP archives; Technical report available Send submissions, questions, address maintenance and requests for old issues to "neuron-request@hplabs.hp.com" or "{any backbone,uunet}!hplabs!neuron-request" Use "ftp" to get old issues from hplpm.hpl.hp.com (15.255.176.205). ------------------------------------------------------------ Subject: tech rep available on evolving neural networks From: Dr Michael G Dyer <dyer@cs.ucla.edu> Date: Wed, 21 Nov 90 09:35:03 -0800 Evolution of Communication in Artificial Organisms* Gregory M. Werner Michael G. Dyer Tech. Rep. UCLA-AI-90-06 Abstract: A population of artificial organisms evolved simple communication protocols for mate finding. Female animals in our artificial environment had the ability to see males and to emit sounds. Male animals were blind, but could hear signals from females. Thus, the environment was designed to favor organisms that evolved to generate and interpret meaningful signals. Starting with random neural networks, the simulation resulted in a progression of generations that exhibit increasingly effective mate finding strategies. In addition, a number of distinct subspecies, i.e. groups with different signaling protocols or "dialects", evolve and compete. These protocols become a behavioral barrier to mating that supports the formation of distinct subspecies. Experiments with physical barriers in the environment were also performed. A partially permeable barrier allows a separate subspecies to evolve and survive for indefinite periods of time, in spite of occasional migration and contact from members of other subspecies. * To appear in: J. D. Farmer, C. Langton, S. Rasmussen & C. Taylor (Eds.), Artificial Life II, Addison-Wesley, in press. For a copy of the above paper, please send a request for Tech. Rep. UCLA-AI-90-06 to: valerie@cs.ucla.edu ------------------------------ Subject: learning a synaptic learning rule. TR available. From: Yoshua BENGIO <yoshua@HOMER.MACH.CS.CMU.EDU> Date: Wed, 21 Nov 90 19:41:36 -0500 The following technical report is now available by ftp from neuroprose: Bengio Y. and Bengio S. (1990). Learning a synaptic learning rule. Technical Report #751, Universite de Montreal, Departement d'informatique et de recherche operationelle. Learning a synaptic learning rule Yoshua Bengio Samy Bengio McGill University, Universite de Montreal School of Computer Science, Departement d'informatique 3480 University street, et de recherche operationelle, Montreal, Qc, Canada, H3A 2A7 Montreal, Qc, Canada, H3C 3J7 yoshua@cs.mcgill.ca bengio@iro.umontreal.ca An original approach to neural modeling is presented, based on the idea of searching for and tuning, with learning methods, a synaptic learning rule which is biologically plausible, and yields networks capable to learn to perform difficult tasks. This method relies on the idea of considering the synaptic modification rule DeltaW() as a parametric function. This function has local inputs and is the same in many neurons. Its parameters can be estimated with known learning methods. For this optimization, we give particular attention to gradient descent and genetic algorithms. Estimation of these parameters consists of a joint global optimization of (a) the synaptic modification function, and (b) the networks that are learning to perform some tasks, using this function. We show how to compute the gradient of an optimization criteria with respect to the parameters of DeltaW(). Both network architecture and the learning function can be designed within constraints derived from biological knowledge. To avoid that DeltaW() be too specialized, this function is forced to be the same for a large number of synapses, in a population of networks learning to perform different tasks. To enforce efficiency constraints, some of these networks should learn complex mappings (as in pattern recognition). Others should learn to reproduce behavioral phenomena, such as associative conditioning, and neurological phenomena, such as habituation, recovery, dishabituation and sensitization. The architecture of the networks reproducing these biological phenomena can be designed based on well-studied circuits, such as those involved in associations in Aplysia, Hermissenda, or the rabbit eyelid closure response. Multiple synaptic modification functions allow for the diverse types of synapses (e.g. inhibitory, excitatory). Models of pre-, epi- and post-synaptic mechanisms can be used to bootstrap Delta W(), so that it initially consists of a combination of simpler modules, each emulating a particular synaptic mechanism. - --------------------------------------------------------------------------- Copies of the postscript file bengio.learn.ps.Z may be obtained from the pub/neuroprose directory in cheops.cis.ohio-state.edu. Either use the Getps script or do this: unix-1> ftp cheops.cis.ohio-state.edu # (or ftp 128.146.8.62) Connected to cheops.cis.ohio-state.edu. Name (cheops.cis.ohio-state.edu:): anonymous 331 Guest login ok, sent ident as password. Password: neuron 230 Guest login ok, access restrictions apply. ftp> cd pub/neuroprose ftp> binary ftp> get bengio.learn.ps.Z ftp> quit unix-2> uncompress bengio.learn.ps.Z unix-3> lpr -P(your_local_postscript_printer) bengio.learn.ps Or, order a hardcopy by sending your physical mail address to bengio@iro.umontreal.ca, mentioning Technical Report #751. Please do this only if you cannot use the ftp method described above. ------------------------------ Subject: tech. rep. available on evolving trail-following organisms From: Dr Michael G Dyer <dyer@CS.UCLA.EDU> Date: Tue, 27 Nov 90 21:29:39 -0800 Evolution as a Theme in Artificial Life: The Genesys/Tracker System* David Jefferson, Robert Collins, Claus Cooper Michael Dyer, Margot Flowers, Richard Korf Charles Taylor, Alan Wang Abstract Direct, fine-grained simulation is a promising way of investigating and modeling natural evolution. In this paper we show how we can model a population of organisms as a population of computer programs, and how the evolutionarily significant activities of organisms (birth, interaction with the environment, migration, sexual reproduction with genetic mutation and recombination, and death) can all be represented by corresponding operations on programs. We illustrate these ideas in a system built for the Connection Machine called Genesys/Tracker, in which artificial "ants" evolve the ability to perform a complex task. In less than 100 generations a population of 64K "random" ants, represented either as finite state automata or as artificial neural nets, evolve the ability to traverse a winding broken "trail" in a rectilinear grid environment. Throughout this study we pay special attention to methodological issues, such as the avoidance of representational artifacts, and to biological verisimilitude. * To appear in J. D. Farmer, C. Langton, S. Rasmussen and C. Taylor (Eds.), Artificial Life II, Addison-Wesley, in press. For a copy of this tech. rep., please send e-mail to: valerie@cs.ucla.edu requesting "Jefferson et al. -- Evolution as Theme in ALife" Be sure to include your USA mail address. ------------------------------ Subject: Dissertation abstract & Technical report From: menon@cs.utexas.edu Date: Wed, 28 Nov 90 12:34:21 -0600 Technical report announcement: Dynamic Aspects of Signaling in Distributed Neural Systems Vinod Menon Dept. of Computer Sciences Univ. of Texas at Austin Austin, TX 78712 ABSTRACT A distributed neural system consists of localized populations of neurons -- neuronal groups -- linked by massive reciprocal connections. Signaling between neuronal groups forms the basis of functioning of such a system. In this thesis, fundamental aspects of signaling are investigated mathematically with particular emphasis on the architecture and temporal self-organizing features of distributed neural systems. Coherent population oscillations, driven by exogenous and endogenous events, serve as autonomous timing mechanisms and are the basis of one possible mechanism of signaling. The theoretical analysis has, therefore, concentrated on a detailed study of the origin and frequency-amplitude-phase characteristics of the oscillations and the emergent features of inter-group reentrant signaling. It is shown that a phase shift between the excitatory and inhibitory components of the interacting intra-neuronal-group signals underlies the generation of oscillations. Such a phase shift is readily induced by delayed inhibition or slowly decaying inhibition. Theoretical analysis shows that a large dynamic frequency-amplitude range is possible by varying the time course of the inhibitory signal. Reentrant signaling between two groups is shown to give rise to synchronization, desynchronization, and resynchronization (with a large jump in frequency and phase difference) of the oscillatory activity as the latency of the reentrant signal is varied. We propose that this phenomenon represents a correlation dependent non-Boolean switching mechanism. A study of triadic neuronal group interactions reveals topological effects -- the existence of stabilizing (closed loop) and destabilizing (open loop) circuits. The analysis indicates (1) the metastable nature of signaling, (2) the existence of time windows in which correlated and uncorrelated activity can take place, and (3) dynamic frequency-amplitude-phase modulation of oscillations. By varying the latencies, and hence the relative phases of the reentrant signals, it is possible to dynamically and selectively modulate the cross-correlation between coactive neuronal groups in a manner that reflects the mapping topology as well as the intrinsic neuronal circuit properties. These mechanisms, we argue, provide dynamic linkage between neuronal groups thereby enabling the distributed neural system to operate in a highly parallel manner without clocks, algorithms, and central control. To obtain a copy of the technical report TR-90-36 please send $5 in US bank check or international money order payable to "The University of Texas" at the following address: Technical Report Center Department of Computer Sciences University of Texas at Austin Taylor Hall 2.124 Austin, TX 78712-1188 USA - ------------------------------------------------------------------------- Dept. of Computer Sciences email: menon@cs.utexas.edu University of Texas at Austin tel: 512-343-8033 Austin, TX 78712 -471-9572 - ------------------------------------------------------------------------- ------------------------------ Subject: Preprint about scaled conjugate gradient available. From: Martin Moller <fodslett@daimi.aau.dk> Date: Fri, 30 Nov 90 19:48:14 +0100 ************************************************************* ******************** PREPRINT announcement: **************** A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning. Martin F. Moller Computer Science Dept. University of Aarhus Denmark e-mail: fodslett@daimi.aau.dk Abstract-- A supervised learning algorithm (Scaled Conjugate Gradient, SCG) with superlinear convergence rate is introduced. SCG uses second order information from the neural network, but requires only O(N) memory usage. The performance of SCG is benchmarked against the performance of the standard backpropagation algorithm (BP) and several recently proposed standard conjugate gradient algorithms. SCG yields a speed-up at least an order of magnitude relative to BP. The speed-up depends on the convergence criterion,i.e., the bigger demand for reduction in error the bigger the speed-up. SCG is fully automated including no user dependent parameters and avoids a time consuming line search, which other conjugate gradient algorithms use in order to determine a good step size. Incorporating problem dependent structural information in the architecture of a neural network often lowers the overall complexity. The smaller the complexity of the neural network relative to the problem domain, the bigger the possibility that the weights space contains long ravines characterized by sharp curvature. While BP is inefficient on these ravine phenomena, SCG handles them effectively. The paper is available by ftp in the neuroprose directory under the name: moller.conjugate-gradient.ps.Z Any question or comments on this short writing or on the preprint would be very much appriciated. Martin M. ------------------------------ Subject: graduate study in neural networks From: caroly@park.bu.edu Date: Fri, 30 Nov 90 15:13:20 -0500 (please post) *********************************************** * * * GRADUATE PROGRAM IN * * COGNITIVE AND NEURAL SYSTEMS (CNS) * * AT BOSTON UNIVERSITY * * * *********************************************** Gail A.Carpenter & Stephen Grossberg, Co-Directors The Boston University graduate program in Cognitive and Neural Systems offers comprehensive advanced training in the neural and computational principles, mechanisms, and architectures that underly human and animal behavior, and the application of neural network architectures to the solution of outstanding technological problems. Applications for Fall, 1991 admissions and financial aid are now being accepted for both the MA and PhD degree programs. To obtain a brochure describing the CNS Program and a set of application materials, write or telephone: Cognitive & Neural Systems Program Boston University 111 Cummington Street, Room 240 Boston, MA 02215 (617) 353-9481 or send a mailing address to: caroly@park.bu.edu Applications for admission and financial aid should be received by the Graduate School Admissions Office no later than January 15. Applicants are required to submit undergraduate (and, if applicable, graduate) transcripts, three letters of recommendation, and Graduate Record Examination (GRE) scores. The Advanced Test should be in the candidate's area of departmental specialization. GRE scores may be waived for MA candidates and, in exceptional cases, for PhD candidates, but absence of these scores may decrease an applicant's chances for admission and financial aid. Description of the CNS Program: The Cognitive and Neural Systems (CNS) Program provides advanced training and research experience for graduate students interested in the neural and computational principles, mechanisms, and architectures that underly human and animal behavior, and the application of neural network architectures to the solution of outstanding technological problems. Students are trained in a broad range of areas concerning cognitive and neural systems, including vision and image processing; speech and language understanding; adaptive pattern recognition; associative learning and long-term memory; cognitive information processing; self-organization; cooperative and competitive network dynamics and short-term memory; reinforcement, motivation, and attention; adaptive sensory-motor control and robotics; and biological rhythms; as well as the mathematical and computational methods needed to support advanced modeling research and applications. The CNS Program awards MA, PhD, and BA/MA degrees. The CNS Program embodies a number of unique features. Its core curriculum consists of eight interdisciplinary graduate courses each of which integrates the psychological, neurobiological, mathematical, and computational information needed to theoretically investigate fundamental issues concerning mind and brain processes and the applications of neural networks to technology. Each course is taught once a week in the evening to make the program available to qualified students, including working professionals, throughout the Boston area. Students develop a coherent area of expertise by designing a program that includes courses in areas such as Biology, Computer Science, Engineering, Mathematics, and Psychology, in addition to courses in the CNS core curriculum. The CNS Program prepares Ph.D. students for thesis research with scientists in one of several Boston University research centers or groups, and with Boston-area scientists collaborating with these centers. The unit most closely linked to the Program is the Center for Adaptive Systems. The Center for Adaptive Systems is also part of the Boston Consortium for Behavioral and Neural Studies, a Boston-area multi-institutional Congressional Center of Excellence. Another multi-institutional Congressional Center of Excellence focussed at Boston University is the Center for the Study of Rhythmic Processes. Other research resources include distinguished research groups in dynamical systems within the mathematics department; in theoretical computer science within the Computer Science Department; in biophysics and computational physics within the Physics Department; in sensory robotics, biomedical engineering, computer and systems engineering, and neuromuscular research within the Engineering School; and in neurophysiology, neuroanatomy, and neuropharmacology at the Medical School. ------------------------------ Subject: TR on Children's Acquisition of Irregular Past Tense available From: Steve Pinker <steve@psyche.mit.edu> Date: Fri, 30 Nov 90 19:07:05 -0500 The following technical report is now available: MIT CENTER FOR COGNITIVE SCIENCE OCCASIONAL PAPER #41 Overregularization Gary F. Marcus Michael Ullman Steven Pinker Michelle Hollander T. John Rosen Fei Xu MIT ABSTRACT Children's overregularization errors such as 'comed' bear on three issues: "U"-shaped development where children get worse over time because of an interaction between memory and rule-governed processes; the unlearning of grammatical errors in the absence of parental negative feedback; and whether cognitive processes are computed by rules or by parallel distributed processing (connectionist) networks. We remedy the lack of quantitative data on overregularization by exhaustively analyzing the 11,500 irregular past tense utterances in the transcribed spontaneous speech of 69 children, and by reviewing the naturalistic and experimental literature. We found: (1) overregularization errors are relatively rare (median 2.5% of irregular past tense forms), suggesting that there is no qualitative defect in children's grammars that must be unlearned. (2) Overregularization occurs at a roughly constant low rate from the late two's into the school-age years, affecting most irregular verbs. (3) Though there is no stage where overregularization errors predominate, one other aspect of U-shaped development was confirmed: an extended period of correct performance before the first overregularization. (4) No support was found for Rumelhart & McClelland's (1986) hypothesis that overregularization is caused by increases in the number or proportion of regular verbs in the input to the past tense system (either parents' tokens, children's tokens, or children's types). Thus the traditional account in which a memory system operates before a rule system cannot be replaced by a connectionist alternative in which a single network displays rotelike or rulelike behavior in response to changes in input statistics. (5) The onset of overregularization is best predicted by the onset of *obligatoriness*: the errors appear when children stop leaving verbs in past tense contexts unmarked (e.g., 'Yesterday I come'). (6) The more often a parent uses an irregular past tense form of a verb, the less often the child overregularizes it. (7) Verbs are protected from overregularization by neighborhoods of similar-sounding irregulars, but are not attracted to overregularization by neighborhoods of similar-sounding regulars. This suggests that the associative properties of connectionist networks may help explain performance with irregulars (via the memory system in which they are stored) but not with regulars. A simple hypothesis explains these phenomena. Children, like adults, obligatorily mark tense, using one of two mechanisms: memory for irregulars, and an affixation rule that can generate a regular past tense form for any verb. Retrieval of an irregular blocks the rule, but children's memory traces for irregulars are not strong enough to guarantee perfect retrieval. When retrieval fails, the rule is applied, and overregularization results. - --------------------------------------------------------------------------- Copies of the postscript file overreg.ps.Z may be obtained electronically from psyche.mit.edu as follows: unix-1> ftp psyche.mit.edu (or ftp 18.88.0.85) Connected to psyche.mit.edu. Name (psyche:): anonymous 331 Guest login ok, sent ident as password. Password: yourname 230 Guest login ok, access restrictions apply. ftp> cd pub 250 CWD command successful. ftp> binary 200 Type set to I. ftp> get overreg.ps.Z 200 PORT command successful. 150 Opening data connection for overreg.ps.Z (18.88.0.154,1500) (253471 bytes). 226 Transfer complete. local: overreg.ps.Z remote: overreg.ps.Z 253471 bytes received in 4.1 seconds (61 Kbytes/s) ftp> quit unix-2> uncompress overreg.ps.Z unix-3> lpr -P(your_local_postscript_printer) overreg.ps Or, order a hardcopy by sending your physical mail address to Eleanor Bonsaint (bonsaint@psyche.mit.edu), asking for Occasional Paper #41, Please do this only if you cannot use the ftp method described above. ------------------------------ Subject: IJCNN-91-Seattle paper deadline is coming up! From: Don Wunsch <dwunsch@blake.u.washington.edu> Date: Mon, 03 Dec 90 11:47:46 -0800 JCNN '91 Seattle Call for Papers The International Neural Networks Society (INNS) and the Institute for Electronic and Electrical Engineers (IEEE) invite all persons interested in the field of Neural Networks to submit papers for possible presentation at the Conference. Papers must be RECEIVED by February 1, 1991. Submissions received after February 1, 1991 will be returned unopened. All submissions will be acknowledged by mail. International authors should submit their work via Air Mail or Express courier so as to ensure timely arrival. Eight copies (one original and seven copies) are required for submission. Do not fold or staple the original, camera-ready copy. Do not number the pages on the front of the camera-ready copy. Papers of no more than six pages, including figures, tables and references, should be written in English, and only complete papers will be considered. Papers must be submitted camera-ready on 8 1/2" by 11" white bond paper with 1" margins on each of the top, bottom, left and right sides, and un-numbered. They should be prepared by a typewriter or letter-quality printer in one-column format, single-spaced, in Times or similar style font of 10 point or larger, and should be printed on one side of the paper only. FAX submissions are not acceptable. Centered at the top of the first page should be the complete title, author name(s), affiliation(s) and mailing address(es). This is followed by a space and then the abstract, up to 15 lines, followed by the text. In an accompanying letter, the fillowing must be included: Corresponding Author: Name, mailing address, telephone and FAX numbers Technical Area (Neurobiology, applications, electronic implementations, optical implementations, image processing, vision, speech, network dynamics, optimization, robotics and control, learning and generalization, neural network architectures, or other) Presentation Format Preferred: Oral or Poster Presenter: Name, mailing address, telephone and FAX numbers If an oral talk is requested, include any special audio/video requests. Special audio/video requests beyond 35mm slide and overhead transparency projectors will be honored only if there are sufficient requests to justify them. If you have special audio/video needs, please contact Sarah Eck at conference management for more information. Send Papers to: IJCNN '91 Seattle Conference Management, Attn. Sarah Eck, MS/GH-22, Suite 108 5001 25th Ave. NE, Seattle, WA 98195 Tel (206) 543-0888 FAX (206) 685-9359 DEADLINE FEBRUARY 1, 1991 Submissions received after this date will be returned unopened. The cost for IJCNN-91-Seattle is $195.00 for INNS and IEEE members for early registration, deadline March 1, 1991. Non-members early registration is $295.00 and students will pay $50.00. Late registration will be honored until June 1, 1991 to members at $295.00, non-members at $395.00, and students at $75.00. On-site registration will be $395.00, $495.00 and $95.00 respectively. Tutorials will be offered at an additional cost of $195.00, or $295.00 for tutorial registration on site. Exhibitors will present the latest in neural networks, including neurocomputers, VLSI neural networks, implementations, software systems and applications at IJCNN-91-SEATTLE. IJCNN-91-SEATTLE is the neural network industry's largest trade show. Hope to see you there! Don ------------------------------ Subject: reprints available From: Dr Michael G Dyer <dyer@CS.UCLA.EDU> Date: Mon, 03 Dec 90 20:23:12 -0800 reprints available: Dyer, M. G. Distributed symbol formation and processing in connectionist networks. Journal of Experimental and Theoretical Artificial Intelligence. Vol. 2, 215-239, 1990. Abstract: Distributed connectionist (DC) systems offer a set of processing features which are distinct from those provided by traditional symbol processing (SP) systems. In general, the features of DC systems are derived from the nature of their distributed representations. Such representations have a microsemantics -- i.e. symbols with similar internal representations tend to have similar processing effects. In contrast, the symbols in SP systems have no intrinsic microsemantics of their own; e.g. SP symbols are formed by concatenating ASCII codes that are static, human engineered, and arbitrary. Such symbols possess only a macrosemantics -- i.e. symbols are placed into structured relationships with other symbols, via pointers, and bindings are propagated via variables. The fact that DC and SP systems each provide a distinct set of useful features serves as a strong research motivation for seeking a synthesis. What is needed for such a synthesis is a method by which symbols can dynamically form their own microsemantics, while at the same time entering into structured, recursive relationships with other symbols, thus developing also a macrosemantics. Here, we describe a general method, called symbol recirculation, for allowing symbols to form their own microsemantics. We then discuss three techniques for implementing variables and bindings in DC systems. Finally, we describe a number of DC systems, based on these techniques, which perform a variety of high-level cognitive tasks. requests for reprints should be sent to: valerie@cs.ucla.edu ------------------------------ Subject: left-out info: IJCNN-91-Seattle From: Don Wunsch <dwunsch@blake.u.washington.edu> Date: Mon, 03 Dec 90 23:29:44 -0800 >Hi, >The date of the conference seems to be missing. Would you please post it on >the network again? Thanks. >Regards, >Dr. Dit-Yan Yeung Thanks, Dr. Yeung, for pointing out my oversight. I don't want to use up all my permitted postings, so I'll just post the few most critical lines, this time including the date. JCNN '91 Seattle, July 8-12, 1991 Papers must be RECEIVED by February 1, 1991. Send Papers to: IJCNN '91 Seattle Conference Management, Attn. Sarah Eck, MS/GH-22, Suite 108 5001 25th Ave. NE, Seattle, WA 98195 Tel (206) 543-0888 FAX (206) 685-9359 The cost for IJCNN-91-Seattle is $195.00 for INNS and IEEE members for early registration, deadline March 1, 1991. Non-members early registration is $295.00 and students will pay $50.00. Late registration will be honored until June 1, 1991 to members at $295.00, non-members at $395.00, and students at $75.00. On-site registration will be $395.00, $495.00 and $95.00 respectively. Finally, another addition: anyone interested in volunteering should contact me at dwunsch@blake.u.washington.edu as soon as you can. Don ------------------------------ Subject: FTP archives; Technical report available From: David Chalmers <dave@cogsci.indiana.edu> Date: Tue, 04 Dec 90 19:17:03 -0500 (1) Following many requests, the bibliography that I have compiled on the philosophy of mind/cognition/AI is now available by anonymous ftp from cogsci.indiana.edu (129.79.238.6). It is contained in 5 files chalmers.bib.* in the directory "pub". Also contained in this archive are various articles by members of the Center for Research on Concepts and Cognition. Instructions for retrieval are given below. (2) The following technical report is now available. THE EVOLUTION OF LEARNING: AN EXPERIMENT IN GENETIC CONNECTIONISM David J. Chalmers Center for Research on Concepts and Cognition Indiana University CRCC-TR-47 This paper explores how an evolutionary process can produce systems that learn. A general framework for the evolution of learning is outlined, and is applied to the task of evolving mechanisms suitable for supervised learning in single-layer neural networks. Dynamic properties of a network's information-processing capacity are encoded genetically, and these properties are subjected to selective pressure based on their success in producing adaptive behavior in diverse environments. As a result of selection and genetic recombination, various successful learning mechanisms evolve, including the well-known delta rule. The effect of environmental diversity on the evolution of learning is investigated, and the role of different kinds of emergent phenomena in genetic and connectionist systems is discussed. A version of this paper appears in _Proceedings of the 1990 Connectionist Models Summer School_ (Touretzky, Elman, Sejnowski and Hinton, eds). - ----------------------------------------------------------------------------- This paper may be retrieved by anonymous ftp from cogsci.indiana.edu (129.79.238.6). The file is chalmers.evolution.ps.Z, in the directory pub. To retrieve, do the following: unix-1> ftp cogsci.indiana.edu # (or ftp 129.79.238.6) Connected to cogsci.indiana.edu Name (cogsci.indiana.edu:): anonymous 331 Guest login ok, sent ident as password. Password: [identification] 230 Guest login ok, access restrictions apply. ftp> cd pub ftp> binary ftp> get chalmers.evolution.ps.Z ftp> quit unix-2> uncompress chalmers.evolution.ps.Z unix-3> lpr -P(your_local_postscript_printer) chalmers.evolution.ps The file is also available from the Ohio State neuroprose archives by the usual methods. If you do not have access to ftp, hardcopies may be obtained by sending e-mail to dave@cogsci.indiana.edu. ------------------------------ End of Neuron Digest [Volume 6 Issue 71] ****************************************