Genetic Algorithms Digest Tuesday, July 17, 2001 Volume 15 : Issue 25 - Send submissions (articles) to GA-List@gmu.edu. DO NOT send submissions to the listproc@gmu.edu address. - To subscribe send email to listproc@gmu.edu containing the following text in the body of the message: subscribe ga-list - To unsubscribe send email to listproc@gmu.edu containing the following text in the body of the message: unsubscribe ga-list - To change your email address, simply unsubscribe the old address and subscribe the new one. - Send other administrative requests to GA-List@gmu.edu. - You can access back issues, GA code, conference announcements, etc., either through the WWW at http://www.aic.nrl.navy.mil/galist/ or through anonymous ftp at ftp.aic.nrl.navy.mil in /pub/galist. -------------------------------- Today's Topics: - Random number generator - Position Announcement: Research Assistant - CFP: IJCAI-01 ML5 - Workshop on Performance Enhancement in KDD - EvoNet Summer School - ACDM2002 - CfP: SAGA 2001, 1st Symposium on Stochastic Algorithms,... - BOOK: Wavelets in Soft Computing - RMI and GA/GP - Re: GP: RMI and GA/GP - Re: RMI and GA/GP - Re: GP: Re: RMI and GA/GP - Re: RMI and GA/GP - GP: GA article in JDJ - Studentships in Natural Computation - CFP: The Inaugural workshop on Artificial Life (AL'01), - An IJCAI-01 paper and its demo -------------------------------- CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference) SCI2001 Evolvable Sys. and Gen. Prog., Orlando, FL USA Jul 22-25, 01 (v15n8) IJCAI-01 WS on Empirical MEthods in AI, Seattle, USA Aug 4, 01 (v15n5) EvoNet Summer School, Thessaloniki, Greece Aug 27-Sept 1, 01 (v15n25) IDAMAP2001 Intelligent Data Analysis in Medicine & Phar Sep 4, 01 (v15n3) FUZZY DAYS Int Conf on Comp Intell, Dortmund, Germany Oct 1-3, 01 (v14n17) ICES2001 4th Int Conf on Evolvable Systems, Tokyo Oct 3-5, 01 (v14n19) MCCS 2001 6th Int. Conf on Msr & Ctrl in Compl. Sys, Ukr Oct. 8-12, 01 (v15n10) IAT2001 2nd Asia Pac Conf on Intell Agent Tech, Japan Oct 23-26, 01 (v14n14) EA01 ÉVOLUTION ARTIFICIELLE 2001, Le Creusot, France Oct 29-31, 01 (v15n5) ICDM01 IEEE Int Conf on Data Mining, Silicon Valley, Nov 29-Dec 2, 01 (v14n14) ANNIE 2001 Smart Eng. Systems Design Conf, StL, MO, USA Nov 4-7, 01 (v15n5) FUZZ-IEEE01 10th IEEE Int Conf on Fuzzy Systems, Austr Dec 2- 5, 01 (v14n20) AL'01 1st Workshop on Art. Life Adelaide, Australia Dec 11, 01 (v15n25) SAGA2001 1st Symp on Stoch Alg..., Berlin, Germany Dec 13-14, 01 (v15n25) AMT01 6th Int Conf Active Media Tech, Hong Kong, China Dec 18-20, 01 (v15n16) NF2002 1st Int ICSC Congress on Neuro-Fuzzy, Cuba Jan 15-18, 02 (v14n18) ICAIS2002 1st ICSC on Aut. Int. Sys., Geelong, Australia Feb 12-15, 02 (v15n22) EVOLANG2002 4th Int Conf on Evolution of Language, USA Mar 27-30, 02 (v15n21) CEC2002 Congress on Evolutionary Comput., Honolulu, HI May 12-17, 02 (v15n23) PATAT 2002 4rth Int. Conf. ... Auto. Timetbl., Belgium Aug 21-23, 02 (v15n10) PPSN VII 7th Int Conf on Parallel Prob.., Granada, Spain Sep 7-11, 02 (v15n21) ICSC-NAISO 3rd ICSC on Eng of Int. Sys., Malago, Spain Sep 24-27, 02 (v15n24) ACDM2002 5th Int. Conf. on Adaptive Comp...Devon, UK Apr 16-18, 02 (v15n25) Send announcements of other activities to GA-List@gmu.edu -------------------------------- Sender: "H.S.Rai" Subject: Random number generator The Evolutionary Computation uses random number for its all operations. Is EA's performance depends on type of random number generator (RNG) used? If so, which algorithms are consider good and used for GA, ES and EP. Hardeep Singh Rai University of Roorkee - India. -------------------------------- Sender: "Richard J. Povinelli" Subject: Position Announcement: Research Assistant Position Announcement: Research Assistant A graduate Research Assistantship is currently available in the Electrical and Computer Engineering Department at Marquette University. This position will be associated with a recently funded NSF project on "Integration of Stochastic and Dynamical Methods for Speech Technology". The goal of this research is to discover time-domain analysis techniques using dynamical systems methods that will lead to improved analysis of speech signals and to improvements in speech recognition accuracy. This goal represents the integration of two traditionally distinct research fields: statistical signal processing and chaotic systems. Since signal processing is fundamentally based on linear systems theory and the study of chaos is inherently non-linear, these fields have little or no overlap outside of the fact that both attempt to model the behavior of physical systems. This research will integrate these very different viewpoints by applying stochastic analysis and modeling tools from the signal processing field to the problem of analyzing embedded phase spaces obtained from chaotic systems analysis of time-series signals. The Speech and Signal Processing lab and Knowledge and Information Discovery lab are associated with the Department of Electrical and Computer Engineering at Marquette, located in Milwaukee, WI (http://www.milwaukee.org). Marquette University is a well-known graduate and undergraduate institution with over 10,000 students and a long history of academic excellence. Interested individuals should send a letter of interest to Dr. Mike Johnson at Mike.Johnson@Marquette.edu or to Dr. Richard Povinelli at Richard.Povinelli@Marquette.edu, as well as applying to the Marquette Graduate School at http://www.grad.mu.edu. Compensation includes tuition reimbursement as well as a $15,000 stipend for a one-year renewable period beginning August 2001 or January 2002. Ideally, applicants will have had previous exposure to one or more of the following areas: signal processing, speech processing, data mining, machine learning, statistical pattern recognition, evolutionary and genetic algorithms, chaos and dynamical systems. Post-doctorate positions are also available. Additional contact information: Dr. Mike Johnson 1515 W. Wisconsin Avenue Milwaukee, WI 53233 (414) 288-0631 Mike.Johnson@Marquette.edu http://www.eng.mu.edu/johnson Dr. Richard Povinelli 1515 W. Wisconsin Avenue Milwaukee, WI 53233 (414) 288-7088 Richard.Povinelli@Marquette.edu http://povinelli.eece.mu.edu -------------------------------- Sender: Kansas State KDD Lab Subject: CFP: IJCAI-01 ML5 - Workshop on Performance Enhancement in KDD CALL FOR PARTICIPATION 15 Jul 2001 IJCAI-2001 Workshop on Wrappers for Performance Enhancement in Knowledge Discovery in Databases (KDD) [workshop code ML-5] http://www.kddresearch.org/KDD/Workshops/IJCAI-2001/ Saturday, 04 Aug 2001 Seattle, Washington, USA WORKSHOP DESCRIPTION The rapidly increasing volume of data collected for decision support applications in commercial, industrial, medical, and defense domains has made it a challenge to scale up knowledge discovery in databases (KDD), the machine learning and knowledge acquisition component of these applications. Many techniques currently applied to KDD admit enhancement through the WRAPPER approach, which uses empirical performance of inductive learning algorithms as feedback to optimize parameters of the learning system. Wrappers include algorithms for performance tuning, especially: optimization of learning system parameters (HYPERPARAMETERS) such as learning rates and model priors; control of solution size; and change of problem representation (or inductive bias optimization). Strategies for changing the representation of a machine learning problem include decomposition of learning tasks into more tractable subproblems; feature construction, or synthesis of more salient or useful input variables; and feature subset selection, also known as variable elimination (a form of relevance determination). This workshop will explore current issues concerning wrapper technologies for KDD applications. WORKSHOP AUDIENCE This workshop is intended for researchers in the area of machine learning, including practitioners of knowledge discovery in databases (KDD) and statistical and computational learning theorists. Intelligent systems researchers with an interest in high-performance computation and large-scale, real-world applications of data mining (e.g., inference and decision support) will also find this workshop of interest. INVITED TALKS "An Intelligent Assistant for the Knowledge Discovery Process" Abraham Bernstein and Foster Provost New York University "Bagging Considered Harmful" Charles Elkan University of California, San Diego ACCEPTED PAPERS "Unsupervised Model Selection via Evolutionary Local Search" Yongseok Kim, W. Nick Street, and Filippo Menczer "Parallel Online Continuous Arcing and a New Framework for Wrapping Parallel Ensembles" Jesse A. Reichler and Harlan D. Harris "Wrapper-Based Feature Selection for Multivariate Leaf Models" Edwin Pednault and Ramesh Natarajan "Wrappers for Automatic Parameter Tuning in Multi-Agent Optimization by Genetic Programming" William H. Hsu and Steven M. Gustafson "A Filter Implementation Using a Committee Machine of Wrappers" Cecil P. Schmidt CALL FOR PARTICIPATION Any interested attendees are welcome. To be invited to the workshop, send e-mail to the organizing committee at: workshop-ijcai2001@www.kddresearch.org For the workshop agenda, electronic version of accepted papers, invited talks, and slide presentations, as well as up-to-date information on the committees and invited speakers, please visit the workshop web site: http://www.kddresearch.org/KDD/Workshops/IJCAI-2001/ IMPORTANT DATES Full Papers due: Monday, 02 April 2001 (extended deadline) Short Papers due: Friday, 06 April 2001 acceptance notification: Monday, 09 April 2001 camera-ready copy due: Friday, 20 April 2001 (extended deadline) workshop Saturday, 04 Aug 2001 ORGANIZING COMMITTEE William H. Hsu (primary contact) Kansas State University Hillol Kargupta University of Maryland Baltimore County Huan Liu Arizona State University Nick Street The University of Iowa ======================================================= William H. Hsu, Ph.D. Assistant Professor of CIS, Kansas State University Director, Lab for Knowledge Discovery in Databases bhsu@cis.ksu.edu, bhsu@ncsa.uiuc.edu http://www.kddresearch.org ICQ: 28651394 ======================================================= -------------------------------- Sender: "Mij Kelly" Subject: EvoNet Summer School (Apologies for cross posting) EvoNet Summer School 27 Aug - 1 Sep 2001 Thessaloniki, Greece http://evonet.dcs.napier.ac.uk/summerschool2001/ The EvoNet Summer School is aimed at PhD students, postdocs, researchers and industry practitioners who want to benefit from using evolutionary computing techniques. The programme follows the successful format of last year's highly successful CoIL Summer School. Tutorials will be given at introductory and at more advanced levels by some of the leading evolutionary computing scientists in Europe. Each research senior will also present a problem and outline some possible solutions, and the Summer School participants will be asked to join together in teams of 4 to tackle their preferred problem. Initially the senior who formulated the problem will work with the team(s) choosing that problem to provide more background info, suggestions, pointers etc, advising them but also allowing the team to determine their own solution. By the end of the week, each team will have prepared a draft paper describing their work and the results, and they will give a 20 minute presentation to the other participants. An atmosphere of co-operative learning between the team members was seen as a major benefit of last year's Summer School, and several draft papers prepared by the teams during the week have now been accepted for publication in academic journals and conference proceedings. A member of the team who won the Best Paper Award last year, described his experience, "Last year I worked and sweated for a whole week, talking with people that think like me, and we produced something very quickly. This year I presented that paper at the EuroGP2001 conference. I cannot imagine having spent that particular week in a better way." For more information, please visit: http://evonet.dcs.napier.ac.uk/summerschool2001/ ========================== Mij Kelly EvoNet - the European Network of Excellence in Evolutionary Computing Visit EvoWeb -- the online information service for anyone with an interest in evolutionary computing: http://evonet.dcs.napier.ac.uk -------------------------------- Sender: "Ian Parmee" Subject: ACDM2002 We are pleased to announce the first call for papers for the: FIFTH INTERNATIONAL CONFERENCE ON ADAPTIVE COMPUTING IN DESIGN AND MANUFACTURE (ACDM 2002) APRIL 16th - 18th, 2002 UNIVERSITY OF EXETER, DEVON, UK http://ww.ad-comtech.co.uk/ACDM2002.htm Sponsored by the UK EPSRC Engineering Network in Adaptive Computing in Design and Manufacture and the University of the West of England CONFERENCE CHAIR: Professor Ian Parmee, University of the West of England, UK [ ... modified by moderator for brevity ... ] THE CONFERENCE This fifth evolutionary / adaptive computing in design and manufacture conference, sponsored by the EPSRC Engineering Network in Adaptive Computing in Design and Manufacture and the University of the West of England, will be held on Exeter University's campus in the South West of England from April 16th-18th, 2002. As in previous years, the intention is to explore the integration of evolutionary / adaptive search, exploration and optimisation and associated Computational Intelligence (CI) technologies (e.g. neural computing, intelligent agent systems, fuzzy logic etc) across a wide spectrum of design and manufacturing activities. Areas of specific interest include but are not restricted to: * the development and integration of appropriate evolutionary and adaptive computing strategies with conceptual, embodiment and detailed design. * novel application of appropriate evolutionary / adaptive computing strategies to or integration with complex manufacturing systems. * scheduling and planning; facility lay-out; supply chain design and management; optimisation of organisational structure. * co-operative frameworks supporting the utilisation of evolutionary / adaptive search and other CI technologies within a design / manufacturing environment. * the application of novel adaptive computing techniques and strategies that address specific design / analysis problems of high complexity. * evolutionary and adaptive strategies for component modelling and systems identification. * design search and exploration; human-centred aspects and interactive evolutionary decision-support systems. * modelling of and searching across, uncertain / poorly-defined decision-making environments. * multi-objective satisfaction and optimisation. * search and optimisation within heavily constrained domains. * Web-based evolutionary tools for design and manufacture - accessibility and utility. * provision of multi-disciplinary search and optimisation requirements within distributed Problem Solving Environments (PSEs). * reducing computational expense during detailed design and analysis. * best practice re integration with high-performance computing, parallel architectures etc * supporting innovative and creative design . * development and integration of aesthetic fitness measures. * data mining; the identification of optimal design information; appropriate presentation of data generated from evolutionary search, exploration and optimisation. * agent-assisted evolutionary design and manufacture processes. * evolvable hardware design. * evolutionary and adaptive computing in manufacturing robotics. Applied, theoretical, results-oriented and speculative papers from both academia and industry will all be considered for inclusion. Application papers must exhibit novel aspects relating to evolutionary / adaptive system design and implementation. The conference retains its single stream format with extended poster sessions and is supported by a number of international invited speakers all of whom are leaders in the field. The format is well suited to a highly interactive meeting where fundamental problems relating to appropriate research and development of these technologies within the design and manufacture domain can be addressed, discussed and resolved. The conference venue is in the attractive grounds of Exeter University's campus which is within walking distance of the city's historic centre (www.exeter.uk.net). SUBMISSION OF PAPERS Papers must be original and related to the overall subject area of the Conference. Initial submissions should be no longer than six pages (A4) at 10pt Times New Roman typeface. Please include author's names, addresses (email included) and affiliation. Margins of 15mm should be maintained all round but overall format is flexible in the first instance. All preliminary papers should be submitted electronically in pdf or postscript format. Accepted papers will receive a layout guide for the preparation of final camera-ready papers. The Scientific Committee will review all papers and those selected for full presentation will again be published in book form by Springer-Verlag. Those selected for the poster session presentations will be included in the ACDM poster proceedings. Important dates: 25th October 2001 Submission of paper 30th December 2001 Notification of acceptance 20th January 2002 Camera-ready copy required Any queries or early submissions should be directed to Ian Parmee on: iparmee@ad-comtech.co.uk -------------------------------- Sender: Kathleen Steinhoefel Subject: CfP: SAGA 2001, 1st Symposium on Stochastic Algorithms, Foundations and Applications ******************************************************************* Sorry if you receive this message more than once ******************************************************************* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Second Call for Papers SAGA 2001 1st Symposium on Stochastic Algorithms, Foundations and Applications (Symposium Stochastische Algorithmen, Grundlagen und Anwendungen) Berlin, December 13-14, 2001 Organized by GMD -- German National Research Center for Information Technology, FIRST -- Research Institute for Computer Architecture and Software. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Scope: The first Symposium on Stochastic Algorithms, Foundations and Applications (SAGA'01) aims at providing a forum for the presentation of original research in the analysis, implementation, experimental evaluation, and real-world application of stochastic algorithms. It focuses, in particular, on new algorithmic ideas involving stochastic decisions and designing and testing stochastic algorithms for realistic environments and scenarios. Therewith, the symposium wants to contribute to fostering the cooperation among practitioners and theoreticians and among algorithmic and complexity researchers of the field. Topics: Original research papers (including significant work-in-progress and work identifying and exploring directions of future research) or state-of-the-art surveys are solicited in all aspects of algorithms employing stochastic components, including, but not limited to: - Stochastic algorithms for combinatorial optimization - Stochastic local search methods - Stochastic machine-learning methods - Run-time analysis and speed of convergence - Average-case behavior and experimental analysis - Parallel and network algorithms - Stochastic complexity results - Various applications Sponsors: DaimlerChrysler AG, GMD FIRST, IT Service Omikron GmbH Venue: The symposium takes place at the Berlin-Brandenburg Academy of Sciences and Humanities, Jaegerstrasse 22/23, 10117 Berlin (near Gendarmenmarkt). http://www.bbaw.de/index_e.html Proceedings: The proceedings will be published in the series Lecture Notes in Computer Science, Springer-Verlag. Submissions: Contributions should consist of a cover page, with the author's full name, address, fax number, e-mail address, a 100-word abstract and keywords (for electronic submission, this page should be in the form of a separate ascii-only e-mail); an extended abstract describing original research in no more than 10 pages, using the Springer LNCS style (available from http://www.springer.de/comp/lncs/authors.html); an optional appendix, with more details to be read/consulted at the discretion of the Program Committee. Submissions should be made by e-mailing a postscript file to saga01@first.gmd.de It is expected that accepted papers will be presented at the conference. Simultaneous submission to other conferences with published proceedings is not allowed. For physical submissions, send 10 copies of the extended abstract to Kathleen Steinhoefel GMD FIRST, Kekulestr. 7, 12489 Berlin, Germany Important Dates: Submission deadline August 3, 2001 Notification to authors September 17, 2001 Final version due October 8, 2001 Symposium December 13-14, 2001 [ ... modified by moderator for brevity ... ] Local Contact: SAGA 2001 GMD FIRST, Kekulestr. 7, 12489 Berlin, Germany Phone: +49 30 6392 1800 Fax: +49 30 6392 1805 Email: saga01@first.gmd.de Further information can be found at http://www.first.gmd.de/saga01/ -------------------------------- Sender: "Thuillard Marc" Subject: BOOK: Wavelets in Soft Computing BOOK ANNOUNCEMENT: WAVELETS IN SOFT COMPUTING by Marc Thuillard World Scientific Series in Robotics and Intelligent Systems - Vol. 25 World Scientific Publishing (Singapore). ISBN 981-02-4609-9, 248pp Pub. date: Jun 2001, hardcover, $48. Further Informations: http://www.wspc.com.sg/books/ or Marc.thuillard@bluemail.ch This book presents the state of integration of wavelet theory and multiresolution analysis into soft computing. It is the first book on hybrid methods combining wavelet analysis with fuzzy logic, neural networks or genetic algorithms. Much attention is given to new approaches (fuzzy-wavelet) that permit one to develop, using wavelet techniques, linguistically interpretable fuzzy systems from data. The book also introduces the reader to wavelet-based genetic algorithms and multiresolution search. A special place is given to methods that have been implemented in real world applications, particularly the different techniques combining fuzzy logic or neural networks with wavelet theory. Contents: * Introduction to Wavelet Theory * Pre-Processing: The Multiresolution Approach * Spline-Based Wavelets Approximation and Compression Algorithms * Automatic Generation of a Fuzzy System with Wavelet Based Methods * On-Line Learning * Nonparametric Wavelet-Based Estimation and Regression Techniques * Developing Intelligent Products * Genetic Algorithms and Multiresolution Readership: Graduate students, researchers, academics/lecturers and industrialists. Dr. Marc Thuillard Head of Research BELIMO Automation AG Brunnenbachstrasse 1 CH-8304 Hinwil Switzerland Marc.thuillard@belimo.ch -------------------------------- Sender: "C. Setzkorn" Subject: RMI and GA/GP Hello, I am planning to use distributed computing (more precisely RMI - remote message invocation - java) to exploit parallelism in GA/GP. I am planning to use all our student machines, which obviously aren?t used all the time. RMI is especially appealing to me because I can send whole objects individuals) over a network and can execute (evaluate) individuals on every type of machine (Linux/Unix, Windows, Mac ?) due to java?s platform independence. I am wondering if anyone has already used RMI for this purpose? and how did they go about doing it? I am also interested in papers which review the area of GA/GP and parallelism. I would really appreciate any pointers regarding this issue. Many thanks in advance. ~~ All the best Chris *********************************************************** Mr. C. Setzkorn (PhD Student) Department of Computer Science University of Liverpool Chadwick Building, room G45a Peach Street, Liverpool L69 7ZF United Kingdom Email: chris@csc.liv.ac.uk Phone: 0044 151 794 3694 Fax: 0044 151 794 3715 homepage: http://www.csc.liv.ac.uk/~chris -------------------------------- Sender: Sean Luke Subject: Re: GP: RMI and GA/GP Chris, there's been a fair amount of swork done in this area that you might want to look at. I have seen roughly five approaches to distributed parallelism in GA and GP. 0. Multiple threads. If you have a multiple-processor, shared-memory architecture, it's not too tough to set up multiple breeding threads to split up the work in breeding and evaluating a single global population. This can be done without locking in a generational approach; steady-state approaches would commonly require locking. Note that this approach can be used in combination with the other four approaches below. 1. High coupling neighborhood models. Here individuals are laid out in some topology, let's say a 2D grid. The selection of individuals to breed to fill a given grid slot is not determined globally through the population (as is usually done) but instead is based on scanning the local neighborhood around the slot. A common approach is to do a random walk of some distance m away from the slot. Various evolutionary-theory advantages are touted for this approach, but that aside, implementing it efficiently in a beowulf cluster is IMHO a nontrivial endeavor. Such a model is most appropriately adapted to vector machines. 2. Island models. With the advent of beowulf clusters, this has become a popular approach. Here each processor runs its own separate evolution process, and every once in a while individuals migrate from process to process, thus maximizing network bandwidth. 3. Distributed evaluation. This approach, which nicely dovetails with steady-state GA, has a central evolution process which farms out evaluation to multiple processors. If a processor gets yanked from the system, no matter, that individual just gets a bad fitness or is placed back in the queue for reevaluation. An approach like this is ncely suited for the loosely-coupled, no-promises networks such as SETI@home or distributing jobs through java applets in people's web browsers. But it's also appropriate for beowulf clusters, especially if your evaluation time far exceeds your required breeding time. 4. Multistart models. The monte carlo simulation community has all sorts of massively parallel, near-zero coupling models for distributed hillclimbing which can also be adapted for GAs etc. Here the idea is that you run an evoolutionaryy process until you determine you can go no further realistically, then you report your results to the central repository and restart your process (the repository may tell you your new initial trajectory). So basically here you're just running totally decoupled parallel processes. > I am also interested in papers which review the area of GA/GP and > parallelism. ICGA had lots of papers on this topic, but I'd suggest first grabbing the Handbook of Evolutionary Computation (if you can -- it's a massive book) which talks about local neighborhood and island models at some length. In the GP community, which as you might guess has deaqlt a lot with paralleiism, some sources you should check out are: - John Koza, who has long built enormous beowulf clusters to run distributed GP experiments. His latest book, GP-III, discusses some issues there. He does primarily #2 I believe - Bill Langdon, who with Phyllis Chong (I think, Bill correct me please) did a distributed GP approach a-la #3 - lil-gp, which has support for multiple thread support #0 and someone has written PVM support I think (#2). I used lil-gp extensively to do distributed evaluation (#3) in a dedicated server farm for my soccer stuff. lil-gp is in C. - ECJ (the program, not the journal), which supports parallelized breeding and evaluation in multiple threads (#0) and also asynchronous island models (#2) with hooks for the rest. ECJ is in Java. I am not aware of anyone using RMI as their transfer mechanism -- most are using threads and/or sockets. For a complete collection of links to known GP systems in Java, see ECJ's home page at http://www.cs.umd.edu/projects/plus/ec/ecj/ For some good EC toolkits, including NRL's ECKit in Java, see the GMU ECLab home page at http://www.cs.gmu.edu/~eclab/ (not the official site URL). Enough horn-tooting, perhaps somone else could post additional links. Sean -------------------------------- Sender: Bill LANGDON Subject: Re: RMI and GA/GP Dear Chris, You might like to have a look at Phyllis Chong's work. She ran GP populations distributed via Java serialisable objects across the world in the summer of 1998. For example: @InProceedings{chong:1999:jDGPi, author = "Fuey Sian Chong and W. B. Langdon", title = "Java based Distributed Genetic Programming on the Internet", booktitle = "Proceedings of the Genetic and Evolutionary Computation Conference", year = "1999", editor = "Wolfgang Banzhaf and Jason Daida and Agoston E. Eiben and Max H. Garzon and Vasant Honavar and Mark Jakiela and Robert E. Smith", volume = "2", pages = "1229", address = "Orlando, Florida, USA", publisher_address = "San Francisco, CA 94104, USA", month = "13-17 " # jul, publisher = "Morgan Kaufmann", keywords = "genetic algorithms, genetic programming, DGP, Distributed Computing, Java Applet / Application, World Wide Computing, Internet, Servlets", ISBN = "1-55860-611-4", URL = "ftp://ftp.cs.bham.ac.uk/pub/authors/W.B.Langdon/papers/p.chong/DGPposter.ps.gz", size = "1 page", abstract = "A distributed approach for parallelising Genetic Programming (GP) on the Internet is proposed and its feasibility demonstrated with a distributed GP system termed DGP developed in Java. DGP is run successfully across the world over the Internet on heterogeneous platforms without any central co-ordination. The run results and the outcome of an experiment to determine DGP's performance are reported together with a description of DGP.", notes = "GECCO-99, part of banzhaf:1999:gecco99, A joint meeting of the eighth international conference on genetic algorithms (ICGA-99) and the fourth annual genetic programming conference (GP-99) see also chong:1999:jDGPi Phyllis Chong", } Bill W. B. Langdon, Phone +44 20 7679 4436 Computer Science, Fax +44 20 7387 1397 University College, London, Gower Street, London, WC1E 6BT, UK http://www.cs.ucl.ac.uk/staff/W.Langdon GECCO San Francisco 7-11 July http://www.isgec.org/GECCO-2001 GP+EM Journal http://www.wkap.nl/journalhome.htm/1389-2576 -------------------------------- Sender: Juan Julian Merelo Guervos Subject: Re: GP: Re: RMI and GA/GP > Ram Janovski wrote: > > Chris, > I would consider using Jini/Javaspaces instead of plain RMI. Indeed, we did it a couple of years ago. Not a big deal: 40 J. Atienza; M. García; J. González; J. J. Merelo. Jenetic: a distributed, fine-grained, asynchronous evolutionary algorithm using Jini. In M. Grana, editor, FEA (Frontiers of Evolutionary Algorithms) proceedings, 2000. 48 M. García-Arenas; J. G. Castellano; P. A. Castillo; J. Carpio; M. Cillero; J. J. Merelo; A. Prieto; V. Rivas; G. Romero. Speedup measurements for a distributed evolutionary algorithm that uses jini. In Universidad de Granada Depto. ATC, editor, XI Jornadas de Paralelismo, 2000. These articles are available from our web site: http://geneura.ugr.es/~jmerelo/research.html J ~~ jmerelo@geneura.ugr.es | jjmerelo@worldonline.es JJ Merelo | http://geneura.ugr.es/~jmerelo Grupo Geneura ~~~~ Univ. Granada | http://www.geneura.org/ -------------------------------- Sender: Juan Julian Merelo Guervos Subject: Re: GP: RMI and GA/GP You could also try using SOAP: there are many implementations of SOAP for Java, including Apache's SOAP at http://xml.apache.org/soap/index.html. I did an implementation of a parallel GA using SOAP and PERL, and published it, unfortunately in Spanish. If you are interested, I can send the paper and/or code to you. Here's the reference: J. J. Merelo, J.G. Castellano, P.A. Castillo, and G. Romero. Algoritmos genéticos distribuidos usando SOAP. In Actas Jornadas de Paralelismo 2001 The paper is also available online from http://geneura.ugr.es/~jmerelo/research.html J ~~ jmerelo@geneura.ugr.es | jjmerelo@worldonline.es JJ Merelo | http://geneura.ugr.es/~jmerelo Grupo Geneura ~~~~ Univ. Granada | http://www.geneura.org/ -------------------------------- Sender: John Porter Subject: GP: GA article in JDJ For all who may have missed it (I did until today), the January 2001 issue of the Java Developer's Journal has an article on GA in Java. ~~ John Porter -------------------------------- Sender: John A Bullinaria Subject: Studentships in Natural Computation ~~~~~~~~~~~~~~~~~~~~~~~~~ Scholarships/Studentships ~~~~~~~~~~~~~~~~~~~~~~~~~ MSc/PGDip/PGCert in Natural Computation ======================================= http://www.cs.bham.ac.uk/natcomp School of Computer Science The University of Birmingham, UK Starting in October 2001, we are offering an advanced 12 month MSc programme in Natural Computation (i.e. computational systems that use ideas and inspirations from natural biological, ecological and physical systems). This will comprise of six taught modules in Neural Computation, Evolutionary Computation, Molecular and Quantum Computation, Nature Inspired Optimisation, Nature Inspired Learning, and Nature Inspired Design; two mini research projects; and one full scale research project. PGDip and PGCert are available for shorter periods. Part-time studies are welcome. The programme is supported by the EPSRC through its Master's Level Training Packages, and by a number of leading companies. It is open to candidates with a very good honours degree or equivalent qualifications in Computer Science/Engineering or closely related areas. A small number of fully funded EPSRC studentships (scholarships) are available to UK/EU students. A small number of scholarships funded by the European Commission under its ASIA-ITC programme are available to Indian students. Additional financial support from our industrial partners may be available during the main project period. Further details are available from our Web-site at: http://www.cs.bham.ac.uk/natcomp or via email: Admissions@cs.bham.ac.uk Please note that the official closing date for applications is 15th July 2001, though good applications may still be considered after that date. -------------------------------- Sender: "Hussein A. Abbass" Subject: CFP: The Inaugural workshop on Artificial Life (AL'01), Adelaide, Australia. 11-Dec-2001 Dear colleague I would appreciate it very much if you can post our first call for papers to your colleague and students. Best Regards, Hussein. The Inaugural workshop on Artificial Life (AL'01) http://www.cs.adfa.edu.au/~abbass/ALL/ Adelaide, Australia. 11-Dec-2001 To be held within the 14th Australian Joint Conference on Artificial Intelligence (AI'01) http://www.cs.adelaide.edu.au/AI2001/ December 2001 IMPORTANT ******* A special issue in an outstanding journal is being investigated Call For Papers =============== The Inaugural workshop on Artificial Life, (AL'01), will be held in conjunction with the Australian Joint Conference on Artificial Intelligence in December 2001. Usually, the term "Artificial Life" (Alife) is used to delineate systems that exhibit some properties of life. Research in Alife ranges from analyzing and understanding life and nature - at least as we may believe it is - to modeling biological systems or solving biological problems. Alife is a large interdisciplinary research area covering research from computer science, biology, chemistry, physics, engineering, sociology, and psychology. Research in Alife can be generically classified into three main areas: · Biological behavior as a metaphor for computational models; such as Ant Algorithms, Ant Colony Optimization, Marriage in Honey-Bees Optimization, Immune Systems, Neural Networks, Evolutionary Algorithms, forest, and DNA computing. · Computational models that reproduce/duplicate a biological behavior; such as Swarm Intelligence, Simulation, Cellular Automaton · Computational models to solve biological problems; such as Bio-informatics The aim of this workshop is to get together computer scientists, biologists, chemists, engineers, geneticists, physicists, and others, to gain more understanding of the mystery hidden in life and to expose researchers to the recent advances in this fast developing area. Topics of interest include, but not limited to, * Ant Algorithms * Ant Colony Optimization * Applications of ALife technologies * Bioinformatics * Biological agents and robotics * Cellular automaton * Complex systems * Emergence * Evolutionary and adaptive dynamics * Marriage in Honey-bees Optimization * Multi-agent systems * Origin of life * Self-organization * Self-replication * Simulation and synthesis tools and methodologies * Swarm Intelligence * and Other related topics Location ======== University of South Australia, Adelaide, Australia Important Dates =============== Submission of papers: 15-Sep-2001 Notification to authors: 26-Oct-2001 Camera ready format: 09-Nov-2001 Workshop date: 11-Dec-2001 Paper Format and Review Process =============================== Each paper will be reviewed by at least two members of the program committee. After the final decision is made regarding the papers, authors will receive the instructions for the final submission of their papers accompanied with the reports of the reviewers. At least one author is expected to attend at the workshop to present his/her paper. All papers must have a title page that includes a title, a 300-400 word abstract, a list of keywords, the names and addresses of all authors, their email addresses, and their telephone and fax numbers. Publication =========== Accepted papers will be published in the workshop proceedings with an ISBN number. The maximum number of pages is 10. A special issue in an outstanding journal is currently being negotiated. All papers submitted to the workshop will be considered to the special issue. Depending on the number of submissions to the workshop, submission to the special issue will most likely be restricted to the workshop participants. Workshop Chair ============== Hussein A. Abbass University of New South Wales, School of Computer Science, Australian Defence Force Academy Campus, NorthCott Drive, Canberra, ACT2600, Australia. Email: abbass@cs.adfa.edu.au Fax: +61-2-62688581 [ ... modified by moderator for brevity ... ] University of New South Wales, School of Computer Science, Australian Defence Force Academy Campus, NorthCott Drive, Canberra, ACT2600, Australia. Email: abbass@cs.adfa.edu.au http: http://www.cs.adfa.edu.au/~abbass Tel.(M) (+61) 0402212977 Tel.(H) (+61) (2) 62578757 Tel.(W) (+61) (2) 62688158 Fax.(W) (+61) (2) 62688581 -------------------------------- Sender: "zhouzh" Subject: An IJCAI-01 paper and its demo dear colleagues, Below is a paper to be presented at IJCAI-01: Title: Genetic algorithm based selective neural network ensemble Author: Zhi-Hua Zhou, Jian-Xin Wu, Yuan Jiang, and Shi-Fu Chen Abstract: Neural network ensemble is a learning paradigm where several neural networks are jointly used to solve a problem. In this paper, the relationship between the generalization ability of the neural network ensemble and the correlation of the individual neural networks is analyzed, which reveals that ensembling a selective subset of individual networks is superior to ensembling all the individual networks in some cases. Therefore an approach named GASEN is proposed, which trains several individual neural networks and then employs genetic algorithm to select an optimum subset of individual networks to constitute an ensemble. Experimental results show that, comparing with a popular ensemble approach, i.e. averaging all, and a theoretically optimum selective ensemble approach, i.e. enumerating, GASEN has preferable performance in generating ensembles with strong generalization ability in relatively small computational cost. The paper is available at http://cs.nju.edu.cn/people/zhouzh/zhouzh.files/Publication/zwjc_ijcai01.pdf Its demo is available at http://cs.nju.edu.cn/people/zhouzh/zhouzh.files/MLNN_Group/freeware/Gasen.zip Regards Zhihua ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Zhi-Hua ZHOU Ph.d. National Lab for Novel Software Technology Nanjing University Hankou Road 22 Nanjing 210093, P.R.China Tel: +86-25-359-3163 Fax: +86-25-330-0710 URL: http://cs.nju.edu.cn/people/zhouzh/ Email: zhouzh@nju.edu.cn ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -------------------------------- End of Genetic Algorithms Digest ********************************