Genetic Algorithms Digest Wednesday, July 2, 1996 Volume 10 : Issue 28 - Do NOT send email or reply to gadistr@aic.nrl.navy.mil (GA List Moderator) - Send submissions (articles) to GA-List@AIC.NRL.NAVY.MIL - Send administrative (subscribe, unsubscribe, change of address, etc.,) requests to GA-List-Request@AIC.NRL.NAVY.MIL ****************************************************************************** - You can access back issues, GA code, conference announcements, etc., either through the WWW at URL http://www.aic.nrl.navy.mil/galist/ or through anonymous ftp at ftp.aic.nrl.navy.mil [192.26.18.68] in /pub/galist. ****************************************************************************** Today's Topics: - PhD scholarship - SAC'97 - 3 IDSIA papers - New Book ---------------------------------------------------------------------- CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference) ICML96 Intl Conf on Machine Learning, Bari, Italy (v9n54) Jul 3-6, 96 GP96 Genetic Programming Conf, Stanford, CA (v9n9) Jul 28-31, 96 FOGA4 Foundations of Genetic Algorithms, San Diego, CA (v9n47) Aug 3-5, 96 AAAI96 Intl Wrkshp Intelligent Adaptive Agents, Portland (v10n4) Aug 4-8, 96 2NWGA 2nd Nordic Wrkshp on Genetic Algorithms, Finland (v9n56) Aug 19-25, 96 EUFIT96 Intelligent Techniques and Soft Computing, Germany (v9n60)Sep 2-5, 96 SAB96 From Animals to Animats, Cape Cod, Massachusetts (v9n31) Sep 9-13, 96 WCNN96 Session on Evolutionary Algorithms, San Diego (v9n62) Sep 15-20, 96 PPSN96 Parallel Problem Solving from Nature, Berlin (v9n17) Sep 22-27, 96 ICGI96 Colloquium on Grammatical Inference, Montpellier (v9n45) Sep 25-27, 96 ICES96 Conf on Evol Sys from Biology to Hardware, Tsukuba (v10n7) Oct 7-8, 96 SEAL96 Asia-Pacific Conf on Sim Evol And Learning, Korea (v10n13)Nov 9-12, 96 SAC97 Symp on Applied Comp (GAs track), San Jose,CA (v10n28) Feb 28-Mar 2, 97 ICCIN97 Comp Intelligence and Neuroscience, North Carolina (v10n16)Mar 2-5,97 FEA97 1st Intl Wrkshp on Frontiers in EAs, North Carolina (v10n21) Mar 2-5,97 ICANNGA97 Intl Conf on Artificial NNs and GAs, England (v10n16) Apr 1-4, 97 EP97 Conf on Evolutionary Programming, Indianapolis (v10n19) Apr 13-16, 97 GP97 Genetic Programming Conference, Stanford, CA Jul 13-16, 97 ECAL97 4th European Conf on Artificial Life, England (v10n8) Jul 28-31, 97 (Send announcements of other activities to GA-List@aic.nrl.navy.mil) ------------------------------ From: Andy.Keane@soton.ac.uk Date: Thu, 27 Jun 1996 12:02:09 +0100 Subject: PhD scholarship We have a recently awarded PhD scholarship from the EPSRC to work on a project entitled "The Integration of Multi-level, Multi-discipline Design Methods with Modern Stochastic Global Optimizers" The main goals of the work are :- To assess the practicality of using a variety of different design analysis methods simultaneously with modern stochastic optimizers so as to achieve faster, more reliable and more innovative optimizations of designs. To study prototype tools based on this approach in collaboration with practising design staff (n.b., it is not the intention of this study to produce a commercially viable code, rather the aim is to provide inputs to enable suitable specifications for such software to be developed as the subject of other projects). To assess which optimization strategies best deal with this kind of approach and how to use clusters of workstations to implement them. This research project aims to investigate how multi-level, multi-discipline design methods could be used more efficiently by modern stochastic global optimizers. Key to the proposed approach is the idea of using generational optimization methods where individual members of each generation are calculated using a variety of different design codes in parallel on networks of computers. This would allow the use of stochastic optimizers which require many hundreds of design evaluations on problems where fully detailed analyses can take significant amounts of computer time to evaluate each design, but where simplified, but less accurate, calculations can be performed more readily. The scholar will work as part of a team tackling this problem and the scholarship pays UK/European Union fees plus maintenance for up to 3.5 years. Applicants need to satisfy the EPSRC requirement of being resident in the UK (please ask for details). Andy Keane Prof. A.J. Keane, Dept. of Mech. Eng., Southampton University, Highfield, Southampton, SO17 1BJ, U.K. Tel +44-1703-592944 FAX +44-1703-593230 http://www.soton.ac.uk/~ajk/welcome.html ------------------------------ From: rogerw@penguin.mcs.utulsa.edu (Roger Wainwright) Date: Thu, 27 Jun 1996 12:56:47 -0500 Subject: SAC'97 ============================================================= | | | CALL FOR PAPERS | | =============== | | | | 12th ANNUAL ACM SYMPOSIUM ON APPLIED COMPUTING (SAC '97) | | February 28-March 2, 1997 | | Hyatt St. Claire, San Jose, California, U. S. A. | | | | ################################################## | | ++Genetic Algorithms and Optimization Track++ | | ################################################## | | | ============================================================= WWW page: http://www.acm.org/conferences/sac97/ Over the past eleven years, the ACM Symposium on Applied Computing has become a primary forum for applied computer scientists and application developers from around the world to interact and present their work. SAC '97 is sponsored by the ACM Special Interest Groups SIGAda, SIGAPP, SIGBIO, and SIGCUE. ################################################################# +++ SPECIAL TRACK ON GENETIC ALGORITHMS AND OPTIMIZATION (GA) +++ Applications using genetic algorithms, genetic programming and other optimization techniques is one of the major tracks in SAC'97. This particular track has had enormous success in past years, and we anticipate another set of excellent papers for this conference. The purpose of this track is to provide a forum for the interchange of ideas, research, development activities, and applications among academicians and practitioners in the areas related to genetic algorithms and other optimization techniques in applications. GUIDELINES FOR SUBMISSION (must be strictly followed): Original unpublished papers dealing with any aspect of computing applications will be considered for presentation and publication in the SAC'97 Proceedings. Accepted papers will be published in the ACM/SAC'97 Conference Proceedings to be printed by the ACM Press. In order to facilitate the blind external review process, submission guidelines must be strictly adhered to: - Submit 6 copies of your manuscript to the Genetic Algorithms Track Chair, (address is given below). The manuscript must clearly indicate that the paper is to be submitted to the GA Track. - Author names and addresses MUST NOT appear in the body of the paper, self-reference must be in the third person, attribution to the author(s) must be in the form of "author", and bibliographical entries by the author(s) must also be in the form of "author". - The body of the paper should not exceed 4,000 words (approximately 15 double-spaced pages). - A separate cover sheet should be attached to one of the copies, containing the title of the paper, the author(s) and affiliation(s), and the address (including e-mail address and fax number, if available) to which correspondence should be sent. - All submissions must be received by AUGUST 15, 1996. PANELS AND WORKSHOPS Panel proposals are encouraged. Panels and workshops dealing with the state of the art in any aspect of Applied Computing are welcomed. Please send a brief abstract of the topic and the resume/vita of the moderator to the Program Chair by AUGUST 15, 1996. ------------------------------------------------- | | | IMPORTANT DATES | | | | AUGUST 15, 1996 PAPER SUBMISSION | | OCTOBER 15, 1996 AUTHOR NOTIFICATION | | DECEMBER 1, 1996 CAMERA-READY COPY DUE | | FEBRUARY 28, 1997 CONFERENCE BEGINS | | | ------------------------------------------------- SAC-97 GENETIC ALGORITHMS TRACK CHAIR: Roger L. Wainwright Track Editor SAC'97, Genetic Algorithms Department of Mathematical and Computer Sciences University of Tulsa 600 South College Ave. Tulsa, OK 74104-3189 Ph: (918) 631-3143 Fax: (918) 631-3077 rogerw@penguin.mcs.utulsa.edu ------------------------------ From: juergen@idsia.ch (Juergen Schmidhuber) Date: Thu, 27 Jun 96 20:57:02 +0200 Subject: 3 IDSIA papers [ WMS: Although these papers don't appear to relate directly to EAs, one of them is appearing in a book on Evolutionary Computation, so this might be of interest. ] 3 related papers available, all based on a recent, novel, general reinforcement learning paradigm that allows for metalearning and incremental self-improvement (IS). ____________________________________________________________________ SIMPLE PRINCIPLES OF METALEARNING Juergen Schmidhuber & Jieyu Zhao & Marco Wiering Technical Report IDSIA-69-96, June 27, 1996 23 pages, 195 K compressed, 662 K uncompressed The goal of metalearning is to generate useful shifts of inductive bias by adapting the current learning strategy in a "useful" way. Our learner leads a single life during which actions are continually executed according to the system's internal state and current policy (a modifiable, probabilistic algorithm mapping environmental inputs and internal states to outputs and new internal states). An action is considered a learning algorithm if it can modify the policy. Effects of learning processes on later learning processes are measured using reward/time ratios. Occasional backtracking enforces success histories of still valid policy modifications corresponding to histories of lifelong reward accelerations. The principle allows for plugging in a wide variety of learning algorithms. In particular, it allows for embedding the learner's policy modification strategy within the policy itself (self-reference). To demonstrate the principle's feasibility in cases where traditional reinforcement learning fails, we test it in complex, non-Markovian, changing environments ("POMDPs"). One of the tasks involves more than 10^13 states, two learners that both cooperate and compete, and strongly delayed reinforcement signals (initially separated by more than 300,000 time steps). ____________________________________________________________________ A GENERAL METHOD FOR INCREMENTAL SELF-IMPROVEMENT AND MULTI-AGENT LEARNING IN UNRESTRICTED ENVIRONMENTS Juergen Schmidhuber To appear in X. Yao, editor, Evolutionary Computation: Theory and Applications. Scientific Publ. Co., Singapore, 1996 (based on "On learning how to learn learning strategies", TR FKI-198-94, TUM 1994). 30 pages, 146 K compressed, 386 K uncompressed. ____________________________________________________________________ INCREMENTAL SELF-IMPROVEMENT FOR LIFE- TIME MULTI-AGENT REINFORCEMENT LEARNING Jieyu Zhao Juergen Schmidhuber To appear in Proc. SAB'96, MIT Press, Cambridge MA, 1996. 10 pages, 107 K compressed, 429 K uncompressed. A spin-off paper of the TR above. It includes another experiment: a multi-agent system consis- ting of 3 co-evolving, IS-based animats chasing each other learns interesting, stochastic predator and prey strategies. (Another spin-off paper is: M. Wiering and J. Schmidhuber. Solving POMDPs using Levin search and EIRA. To be presented by MW at ML'96.) ____________________________________________________________________ To obtain copies, use ftp, or try the web: http://www.idsia.ch/~juergen/onlinepub.html FTP-host: ftp.idsia.ch FTP-filenames: /pub/juergen/meta.ps.gz /pub/juergen/ec96.ps.gz /pub/jieyu/sab96.ps.gz ____________________________________________________________________ Juergen Schmidhuber & Jieyu Zhao & Marco Wiering http://www.idsia.ch IDSIA ------------------------------ From: matt lybanon Date: Fri, 28 Jun 1996 09:46:41 -0500 Subject: New Book The following book may be of interest to GA List subscribers. Sankar K. Pal and Paul P. Wang (ed.), _Genetic Algorithms for Pattern Recognition_, Boca Raton: CRC Press, 1996, ISBN 0-8493-9467-8 ======================================================================== Matthew Lybanon Remote Sensing Applications Branch Naval Research Laboratory Stennis Space Center, MS 39529 (601) 688-5263 FAX (601) 688-4149 lybanon@snaps.nrlssc.navy.mil ======================================================================== ------------------------------ End of Genetic Algorithms Digest ******************************