Genetic Algorithms Digest Monday, April 19, 1999 Volume 13 : Issue 7 - Do NOT send email or reply to gadistr@aic.nrl.navy.mil - Send submissions (articles) to GA-List@aic.nrl.navy.mil - Send administrative requests (subscribe, unsubscribe, change of address, etc.,) 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 [132.250.84.25] in /pub/galist. ********************************************************************** Today's Topics: - Re: GEC? (Re: v13n5) - Implementation of Evolutionary Strategies - Niching and Co-Evolution combined - Combinatorial optimization with Bayesian networks - Research post available - PhD job announcement - CFP: KES'99 - CFP: Evolutionary Robotics - [New book] Genetic Programming III - EuroGP'99 programme ---------------------------------------------------------------------- CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference) EuroGP99 2nd Euro Workshop on GP, Goteborg, Sweden May 26-29, 99 (v12n20) EvoIASP99 1st Euro WS on EC in Image Anal & Sig Proc May 28, 99 (v12n18) EuroEctel99 1st Euro Workshop on EC in Telecommun May 29, 99 (v12n16) EUROGEN99 Short Course on EAs, Jyvaskyla, Finland May 30-Jun3, 99 (v13n3) SOCO99 Soft Computing, Genova, Italy Jun 1-4, 99 (v12n8) 1999 Complex Systems Summer School Jun 6-Jul 2, 99 (v13n2) ISAMA99 Int Soc for Arts, Math and Arch, Spain Jun 7-11, 99 (v12n21) MENDEL99 5th Int Mendel Conf on Soft Comp, Czech Rep Jun 9-12, 99 (v12n23) TAINN99 8th Turkish Symposium on AI and Neural Nets Jun 23-25, 99 (v12n17) CEF99 EC in Economics and Finance, Chestnut Hill, MA Jun 24-26, 99 (v12n21) ICML99 16th Int Conf on Mach Learning, Bled, Slovenia Jun 27-30, 99 (v12n21) Recent Advances in Soft Computing99, Leicester, UK Jul 1-2, 99 (v12n20) CEC99 Congress on Evol Computation, Washington, DC Jul 6-9, 99 (v12n9) ACAI99 Workshop on ML and Int Agents, Chania, Greece Jul 5-16, 99 (v13n3) GECCO99 Genetic & Evol Computation Conf, Orlando, FL Jul 13-17, 99 (v12n8) AAAI99 16th National Conference on AI, Orlando, FL Jul 18-22, 99 (v12n21) EH99 1st NASA/DOD WS on Evol Hardware, Pasadena, CA Jul 19-21, 99 (v12n23) IJCAI99 WS on Neur, Symb, & Reinfcmt Meth for Sequ Lear Aug 1, 99 (v13n1) IJCAI99 WS on Agents Learning About, From, and With Ot Aug 2, 99 (v12n22) IPCAT99 3rd Int WS on Info Proc in Cells and Tissues Aug 23-26, 99 (v13n3) KES99 3rd Int Conf on KB Intel Inf Eng Sys, Australia Aug 31-Sep 1 99 (v13n7) ECAL99 5th Euro Conf on Artificial Life, Lausanne, Swi Sep 13-17, 99 (v12n20) EUFIT99 7th Euro Conf on Intell Tech and Soft Comp Sep 13-16, 99 (v13n3) SMC99 Int Conf on System, Man, and Cybernetics, Tokyo Oct 12-15, 99 (v13n4) EA99 Artificial Evolution, Dunkerque, France Nov 3-5, 99 (v13n5) ANNIE99 Artificial NN in Engineering, St. Louis, MO Nov 7-10, 99 (v13n5) RSFDGrC99 7th Int WS Rough Sets, Fuzzy Sets, Data Min Nov 9-11, 99 (v12n22) IAT99 Intelligent Agent Technology, Hong Kong Dec 15-17, 99 (v12n21) WAC2000 8th Int Sym on Robotics with Apps, Maui, Hawaii Jun 11-16, 00 (v13n7) Send announcements of other activities to GA-List@aic.nrl.navy.mil. ------------------------------ Date: Mon, 12 Apr 1999 13:31:06 -0400 From: Kenneth DeJong Subject: Re: GEC? (Re: v13n5) I'm not sure who Paul Winchell is either, but I'm a little surprised that the replies to his queries about GECCO and CEC were as lengthy as they were and yet neglected to mention an important and relevant historical event. In 1991 a group of representatives from the GA, ES, and EP communities met in San Diego at ICGA-91 to discuss a shared perception that it was time to call what they were doing a "field", define its scope, provide some cohesive structure, establish a journal, etc. One of the obvious first questions was: what should the name of this field be? It was not an easy question to answer. There were several months of discussions involving all of the leading people in the field and, as I recall, even a broader discussion of this issue on GA-List. The consensus that emerged from all this was the term "Evolutionary Computation" as the succinct description of the intellectual core of the field: a shared interest in the use of Darwinian-like principles of evolution to solve difficult computational problems. That consensus paved the way for the establishment of the journal Evolutionary Computation, the initiation of the ambitious "Handbook of Evolutionary Computation" project, the start of IEEE-related activities including a conference on EC and a Transactions on EC, and many other activities. At the same time I don't believe that anyone meant to imply that consensus on a name also expressed a need for uniformity and homogeneity of the constituents. Historical methodologies like GAs, ES, and EP would continue to expand and develop. New methodologies like GP would emerge. Like economies, countries, etc., a scientific field develops best when it has a healthy blend of cooperation and competition. However, competition, by its nature, requires one to emphasize differences rather than similarities, and results in new labels and new terminology intended to articulate those differences. But "new" is not always the same as "good" or "useful". Like Paul Winchell, I find the term "genetic and evolutionary computation" confusingly redundant. It only makes sense to me if I interpret it to mean "genetic and other forms of evolutionary computation". GOFEC anyone? Kenneth De Jong ------------------------------ Date: Wed, 31 Mar 1999 11:07:23 +0100 From: Joao Carlos Costa Subject: Implementation of Evolutionary Strategies Hi, Does anyone know where can I find a reference implementation of Evolutionary Strategies? I and a couple of coleagues have implemented ES, and have been trying to replicate the results of Prof. Back's book "Evolutionary Algorithms in Theory and Practice" as a test to the library. Since our results are as good as the ones in the book, we would like to look at other implementations, in order to find out what our error might be. Thanks in advance, Joao Costa ------------------------------ Date: Sat, 17 Apr 1999 17:59:59 +0200 From: "Ariel Fligler" Subject: Niching and Co-Evolution combined Greetings All, I was wandering if anyone knows about papers targeting the combination of Niching and Co-Evolution. More specifically, I am interested in niching techniques where the niches are evolving together ( the exitence of other niches is part of each niche fitness value). Any information regarding this topic from any angle will be very helpful. Sorry for the double posting to comp.ai.genetic. Thank you in advance, Ariel Fligler afligler@atl.co.il [ Moderator Note: I happen to have some publications, including a dissertation, on this very topic. Please see my Web page at http://www.cs.gmu.edu/~mpotter for more information. ] ------------------------------ Date: Mon, 19 Apr 1999 14:48:18 +0200 (MET DST) From: Larranaga Mugica Pedro Subject: Combinatorial optimization with Bayesian networks During the last year we have been working on one approach belonging to the EDA (Estimation of Distribution Algorithms) family of algorithms. In our approach, the estimation of the joint probability distribution of selected individuals is done by multiply connected Bayesian networks. These probabilistic graphical models are induced from the data sets of selected individuals in every generation. Recently we have presented part of the work alredady done in two comumunications in the congress CIMAF'99. The titles of the works are: * A review of the cooperation between evolutionary computation and probabilistic graphical models. * Global optimization with Bayesian networks. The postcript versions of the works can be obtained from the web page of the group: http://www.sc.ehu.es/isg Comments welcome ! Prof. Pedro Larranaga ------------------------------ Date: Wed, 07 Apr 1999 17:08:39 +0100 From: Andy Connor Subject: Research post available University of Cambridge, Department of Engineering, Engineering Design Centre RESEACH POSTS Applications are invited for a post at the Research Assistant / Research Associate level to work on the development of multi-objective design optimisation techniques. The appointment is tenable for one year, in the first instance. Further details and an application form may be obtained by writing to Mrs Christa Croghan, Engineering Design Centre, Department of Engineering, Trumpington Street, Cambridge CB2 1PZ or by e-mail cc2@eng.cam.ac.uk. The closing date for applications is 30th April 1999. This is a re-advertisement and previous applicants need not apply. The University is an equal opportunities employer. Andy Connor Research Associate, Cambridge Engineering Design Centre http://www.cus.cam.ac.uk/~amc50/home.html ------------------------------ Date: Thu, 8 Apr 1999 16:31:57 +0200 From: juergen@idsia.ch (Juergen Schmidhuber) Subject: PhD job announcement PHD STUDENT WANTED I am seeking an outstanding PhD candidate for a research project on reinforcement learning (RL) and program evolution (PE). OVERVIEW. Most machine learning research focuses on learning memory-free mappings between input patterns and output patterns. Humans, however, obviously learn entire algorithms mapping input sequences to output sequences in a complex fashion. In particular, they constantly learn to identify important events in input streams and store them in short-term memory until the memories are needed to compute appropriate output actions. If we want to bridge the gap between the learning abilities of humans and machines then we will have to study how such sequential processes can be learned. The focus of this project will be on RL and PE methods whose search space consists of fairly arbitrary, possibly probabilistic "programs" (as opposed to more limited stimulus/response associations). POSSIBLE PROJECT SUBGOALS. The project allows considerable scientific freedom. If you have a great idea, let's go for it and try it. Otherwise we'll start along the following lines. (1) Explore the limits of recent algorithmic search techniques such as "Adaptive Levin Search" and "Probabilistic Incremental Program Evolution" - both of which can learn memory strategies. (2) Improve, extend, and apply a recent technique "Incremental self-improvement" based on the success-story algorithm for probabilistic, self-modifying systems that can in principle learn to improve their own learning algorithm (metalearning). (3) Build unsupervised, "curious" systems selecting their own training exemplars for building models of the environment, and use the models to speed up improvement of goal-directed sequential behavior. (4) Examine RL economies where agents learn to pay each other for useful services, and test whether they can learn to memorize. See http://www.idsia.ch/~juergen/topics.html for papers on the above subjects. A highly qualified candidate is sought with a background in computational sciences, mathematics, engineering, physics or other relevant areas. Applicants should submit : (i) Detailed curriculum vitae, (ii) List of three references (and their email addresses), (ii) Transcripts of undergraduate and graduate (if applicable) studies and (iii) Concise statement of their research interests (two pages max). Candidates are also encouraged to submit their scores in the Graduate Record Examination (GRE) general test (if available). Please send hardcopies of all documents to: Juergen Schmidhuber, IDSIA, Corso Elvezia 36, 6900-Lugano, Switzerland Applications (with WWW pointers to studies or papers, if available) can also be submitted electronically (in plain ASCII or postscript format, but only small files please) to juergen@idsia.ch. Please connect your first and last name by a dot "." in the subject header, and add a meaningful extension. For instance, if your name is John Smith, then your messages could have headers such as: subject: John.Smith.cv.ps, subject: John.Smith.statement.txt, subject: John.Smith.correspondence.txt.... This will facilitate appropriate filing of your stuff. Thanks a lot! ABOUT IDSIA. Our research focuses on artificial neural nets, reinforcement learning, complexity and generalization issues, unsupervised learning and information theory, forecasting, combinatorial optimization, evolutionary computation. IDSIA's algorithms hold the world records for several important operations research benchmarks. In the "X-Lab Survey" by Business Week magazine, IDSIA was ranked in fourth place in the category "COMPUTER SCIENCE - BIOLOGICALLY INSPIRED". Its comparatively tiny size notwithstanding, IDSIA also ranked among the top ten labs worldwide in the broader category "ARTIFICIAL INTELLIGENCE". We are located in the beautiful city of Lugano in Ticino, the scenic southernmost province of Switzerland. Milano, Italy's center of fashion and finance, is 1 hour away, Venice 3 hours. Our collaborators at the Swiss supercomputing center CSCS are nearby; the new University of Lugano is across the lawn. Switzerland (origin of special relativity and the World Wide Web) boasts the highest citation impact factor, the highest supercomputing capacity pc (per capita), the most Nobel prizes pc (450% of the US value), the highest income pc, and perhaps the best chocolate. SALARY: commensurate with experience but generally attractive. Low taxes. There also is travel funding in case of papers accepted at important conferences. Juergen Schmidhuber research director IDSIA, Corso Elvezia 36, 6900-Lugano, Switzerland juergen@idsia.ch http://www.idsia.ch/~juergen ------------------------------ Date: Mon, 8 Mar 1999 17:24:23 +0800 (HKT) From: "Dr. Szeto" Subject: CFP: KES'99 Dear Distinguished Colleagues, I am calling for your contributions in the following International conference in Australia. THIRD INTERNATIONAL CONFERENCE ON KNOWLEDGE-BASED INTELLIGENT INFORMATION ENGINEERING SYSTEMS KES'99. I think that your participation will make it a very successful and exciting meeting. Please forward your papers on evolution computation to me by email or send your hard copies in triplicate at the followig address Prof. K.Y. Szeto Dept. of Physics HKUST, Clear Water Bay Hong Kong e-mail: phszeto@ust.hk Thanks for your attention, K.Y. Szeto --- THIRD INTERNATIONAL CONFERENCE ON KNOWLEDGE-BASED INTELLIGENT INFORMATION ENGINEERING SYSTEMS KES'99 31st August - 1st September 1999 Adelaide Australia The aim of this conference is to provide an international forum for presentation of recent results in the general areas of Intelligent Systems Design and Industrial Applications of Information Technology. The conference will consist of plenary sessions, and contributory sessions mainly on the theory and applications of knowledge-based intelligent information systems using: * KNOWLEDGE-BASED SYSTEMS * ARTIFICIAL NEURAL NETS * FUZZY SYSTEMS * EVOLUTIONARY COMPUTING * CHAOS THEORY * INFORMATION TECHNOLOGY AND MULTIMEDIA SUBMISSION OF PAPERS - Papers must be written in English (4 pages maximum). - Paper presentation is about 15 minutes including questions and discussions. - Include corresponding author with full name, address, telephone and fax numbers, E-mail address. - Include presenter address and his/her 4 line resume for introduction purposes only. - Fax or e-mail copies are not acceptable. - The conference proceedings will be published by IEEE. - Please submit one original and three copies of the camera ready paper (A4 size), two column format in Times New Roman, or similar font style, 10 points with one inch margin on all four sides for review to: L.C. Jain, SCT Building, PES Knowledge-based Intelligent Engineering Systems Centre University of South Australia, Adelaide The Mawson Lakes, S.A., 5095, Australia. e-mail: L.Jain@unisa.edu.au INVITED LECTURES The conference committee is also soliciting proposals for invited sessions focussing on new or emerging electronic technologies. Researchers, application engineers and managers are invited to submit proposals to L.C. Jain by 31 March 1999. DEADLINES Receipt of papers: 14th May 1999 Notification of acceptance: 14th June 1999 FURTHER CONFERNCE DETAILS If you would like your name added to the conference mailing list for registration forms etc, please send your contact details to: Mrs.Kathleen Martin KES99 Secretariat 76 Fifth Avenue Joslin, SA 5070 AUSTRALIA e-mail: petheton@camtech.net.au ------------------------------ Date: Fri, 09 Apr 1999 10:14:54 -0500 From: Gary Parker Subject: CFP: Evolutionary Robotics Call for Papers for an Invited Session on Evolutionary Robotics The Eighth International Symposium on Robotics with Applications World Automation Congress (WAC 2000) http://wacong.com/wac_info.html 11-16 June 2000. Maui, Hawaii, USA. As part of the Eighth International Symposium on Robotics with Applications (ISORA), a special session is planned on evolutionary robotics. The purpose of this session will be to examine alternate means of using the various forms of evolutionary computation to provide learning in robots. Learning methods can be (but are not limited to): Genetic Algorithms Genetic Programming Classifier Systems Evolutionary Strategies Evolutionary Programming Topics of interest include (but are not limited to) the use of these methods to produce: Autonomous Robot Controllers Locomotion Control Motion/Path Planning Obstacle Avoidance Legged Robot Gaits Real-Time Robot Navigation Coordination of Multiple Robots Robot Manipulator Control Emergent Behaviors of Multiple Agents Biologically Inspired Systems Interested researchers should send an abstract of up to 400 words via email or land mail to the address below by 26 April 1999. Invited authors will have full papers published in the symposium proceedings. Gary B. Parker Lindley Hall 215 Computer Science Department Indiana University Bloomington, IN 47405-4101 USA email: gaparker@cs.indiana.edu Important Dates: 26 April 1999: Abstract Due 1 August 1999: Notification of Preliminary Acceptance 1 December 1999: Full Manuscript Due Date 28 February 2000: Final Acceptance 11-16 June 2000: WAC 2000 ------------------------------ Date: Tue, 13 Apr 1999 12:08:12 -0700 From: bemerson@mkp.com Subject: [New book] Genetic Programming III GENETIC PROGRAMMING III Darwinian Invention & Problem Solving by John R. Koza, Forrest H Bennett III, David Andre, and Martin A. Keane Morgan Kaufmann Publishers, April 1999, ISBN 1-55860-543-6 Genetic programming is a method for getting a computer to solve a problem by telling it what needs to be done instead of how to do it. The authors present genetically evolved solutions to dozens of problems of design, optimal control, classification, system identification, function learning, and computational molecular biology. Among the solutions are 14 results competitive with human-produced results, including 10 rediscoveries of previously patented inventions. Researchers in artificial intelligence, machine learning, evolutionary computation, and genetic algorithms will find this an essential reference to the most recent and most important results in the rapidly growing field of genetic programming. FEATURES -- Explains how the success of genetic programming arises from seven fundamental differences distinguishing it from conventional approaches to artificial intelligence and machine learning -- Describes how genetic programming uses architecture-altering operations to make on-the-fly decisions on whether to use subroutines, loops, recursions, and memory -- Demonstrates that genetic programming possesses 16 attributes that can reasonably be expected of a system for automatically creating computer programs -- Presents the general-purpose Genetic Programming Problem Solver -- Includes an introduction to genetic programming for the uninitiated CONTENTS Introduction -- Background -- Architecture-Altering Operations -- Genetic Programming Problem Solver (GPPS) -- Automated Synthesis of Analog Electrical Circuits -- Evolvable Hardware -- Discovery of Cellular Automata Rules -- Discovery of Motifs and Programmatic Motifs for Molecular Biology -- Parallelization and Implementation Issues -- Conclusion MORE INFORMATION http://www.mkp.com/books_catalog/1-55860-543-6.asp Morgan Kaufmann Publishers San Francisco, California http://www.mkp.com orders@mkp.com ------------------------------ Date: Wed, 14 Apr 1999 14:01:59 +0200 From: Bill Langdon Subject: EuroGP'99 programme EuroGP'99, EuroEctel'99, EvoIASP99, EvoRobot'99 GOTEBORG, Sweden 26-29 May This year EuroGP and three other major evolutionary computing workshops will take place in the historic University of Goteborg, Sweden. See http://www.dcs.napier.ac.uk/evonet/Coordinator/html/evoworkshops_99.html EuroGP'99 includes: A tutorial on GP by John Koza An invited talk by David B Fogel Oral and poster sessions with periods for discussion Software demos and industrial stands There is a combined registration where by registering for one workshop you get all of the above plus FREE registration to the other three workshops. In addition there are significant discounts for students See http://146.176.160.175/database/evows1.html for registration details The timetable is 23 Apr 1999 Deadline for EuroGP'99 and EvoISAP late breaking papers http://www.cs.bham.ac.uk/~rmp/eebic/eurogp99/eurogp99_lbp.html http://www.ce.unipr.it/people/cagnoni/evoiasp99_lbp.html 26 May 1999 EuroGP'99 (day 1) 27 May 1999 EuroGP'99 (day 2) 28 May 1999 EvoRobot'99 (day 1) EvoIASP'99 29 May 1999 EvoRobot'99 (day 2) EuroEctel'99 The program for EuroGP'99 can also be found at http://www.cs.bham.ac.uk/~rmp/eebic/eurogp99/programme.html ------------------------------ End of Genetic Algorithms Digest ******************************