Genetic Algorithms Digest Monday, July 24, 2000 Volume 14 : Issue 13 - 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 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: - 1 research programmer and 1 systems programmer positions - Metaheuristics Network: doctoral and postdoctoral positions - WSC5 Virtual Exhibition - PhD Thesis available: MLP Error Surfaces - EPSRC PhD Studentship in AI at City University - FOGA 2000 Program - CSMR 2001 call for papers - MathEngine Evolved Creatures - Deadline extention - ISI'2001 workshop - JOB: Research Fellow, The University of Birmingham, England - PhD Thesis Research / Post-Doc Fellowship for Research in GP - PhD positions at TU Darmstadt, Germany - PPSN 2000 - Paris, France, 16-20 Sept. 2000 - Errata on "Evolutionary optimization ..." in GECCO-2000 Proc. - Special Issue of the Complex Systems Journal - Knowledge and Information Systems: Vol 2 No 3 (2000) - Gene expression programming: a new GP-style technique - Vacant Post: Imperial College of Sci Tech and Medicine - New Classifier Systems Book ------------------------------------------------------------------------------- CALENDAR OF GA-RELATED ACTIVITIES: (with GA-List issue reference) JOCE2000 1st Int Journeys in Evol Comp and Applic, Colom Jul 21-22, 00 (v14n12) IJCNN2000 Int Joint Conf on Neural Networks, Como, Italy Jul 24-27, 00 (v13n23) IAS6 The 6Th Int Conf on Intell Auton Sys, Venice, Italy Jul 25-27, 00 (v13n29) ALIFE VII 7th Int Conf on Artificial Life, Portland, OR Aug 1- 6, 00 (v13n25) PATAT2000 3rd Int Conf Prac & Theor of Auto Timetabling Aug 16-18, 00 (v13n8) ECAI2000 14th Euro Conf on AI, Berlin, Germany Aug 20-25, 00 (v13n26) KES2000 4th Int Conf KB Intell Engr Sys & Allied Tech Aug 30-Sep 1, 00 (v14n4) WSC5 5th Online World Conf on Soft Comp in Indust Apps Sept 4-18, 00 (v14n7) HUMANOIDS2000 1st IEEE-RAS Conf Humanoid Robots, Boston Sep 7-8, 00 (v14n1) ANTS2000 2nd Int WS on Ant Algorithms, Brussels, Belgium Sep 8-9, 00 (v13n28) SAB2000 Int Conf on Sim Of Adaptive Behavior, Paris, FR Sep 11-15, 00 (v13n24) IWLCS2000 3rd Int WS on Learning Class Systems, Paris Sep 16, 00 (v13n26) PPSNVI Parallel Problem Solving from Nature, Paris, FR Sep 16-20, 00 (v13n24) ISMIS00 Int Sym Meth for Intell Sys, SS on EC, Charlotte Oct 11-14, 00 (v14n1) RSCTC2000 2nd Int Conf Rough Sets & Cur Trends in Comp Oct 16-19, 00 (v13n26) SEAL2000 3rd Asia-Pac Conf on Sim Evol and Learning Oct 25-27, 00 (v13n25) ADVIS2000 1st Biennial Int Conf on Adv in IS, Izmir, TR Oct 25-27, 00 (v14n5) IROS2000 Int Conf on Intell Robots and Systems, Japan Oct 30-Nov 5, 00 (v14n7) ANNIE2000 Artificial NN in Engineering, St. Louis, MO Nov 5-8, 00 (v14n4) SCCC2000 XX Int Conf Chilean CS Soc - WS Adv & Trends AI Nov 16-18, 00 (v14n12) SBRN2000 VI Braz Symp Neural Networks, Rio de Janeiro Nov 22-25, 00 (v14n12) ICARCV2000 6th Int Conf on Cont/Aut/Rob/Vis, Singapore Dec 6-8, 00 (v13n28) ISA2000 Int Congress on Intell Sys and Appl, Sydney, AU Dec 12-15, 00 (v14n5) KBCS2000 Int Conf on Knowl Based Comp Sys, Mumbai, India Dec 17-19, 00 (v14n9) EMO01 1st Int Con of Evol Multi-Criterion Opt, Zurich Mar 7-9, 01 (v14n4) CSMR2001 5th Eur Conf on Soft Maint and Reeng, Portugal, Mar 14-16, 01 (v14n13) ISI2001 Int Congress on Info Science Innovations, Dubai Mar 20-23, 01 (v13n25) ICANNGA2001 5th Int Conf on Artif NN and GAs, Prague Apr 22-25, 01 (v14n11) Send announcements of other activities to GA-List@aic.nrl.navy.mil. ------------------------------ Date: Fri, 23 Jun 2000 11:53:39 -0700 From: "John Koza" Subject: 1 research programmer and 1 systems programmer positions 2 IMMEDIATE OPENINGS 1 SYSTEMS PROGRAMMER 1 RESEARCH PROGRAMMER Genetic Programming Inc. is seeking 2 full-time programmers (1 RESEARCH PROGRAMMER and 1 SYSTEMS PROGRAMMER) who are proficient in Java or C. Genetic Programming Inc. is a small privately funded corporation using a beowulf cluster computer system consisting of 1,000 Pentium II processors to do research in applying genetic programming to produce human-competitive results. Our group publishes numerous research papers each year in various scientific conferences and journals. The RESEARCH PROGRAMMER position calls for at least a B.S. degree and at least two years experience in Java or C doing academic or corporate research programming. The successful candidate is expected to have expertise in some specific science or engineering domain with the aim of applying genetic programming to that domain. The possibilities are very open-ended and include, but are not limited to, operations research, finite element analysis, shape optimization, fluid dynamics, mechanical design, aeronautical design, optics, signal processing, civil engineering, applied mathematics, chemical engineering, bioinformatics, computational biology, genomics, protein structures, etc. This position requires the ability to discuss and contribute ideas on how to approach new problems, suggest new problem areas to address, quickly learn about new problem domains, and to then write and run the necessary computers programs. The SYSTEMS PROGRAMMER position calls for at least a B.S. degree, two years experience as software developer in Java or C, and ability to maintain our software and our 1,000-Pentium system. Familiarity with linux, network programming, shell scripting, perl is a plus. The successful candidate will also be actively involved in our research programming. Experience with, or knowledge of, genetic programming, genetic algorithms, or evolutionary computation, machine learning, neural networks, artificial intelligence, and artificial life is a plus. Please include all relevant information, date available, and several references. Since these positions are available now, please be aware of the difficulty, paperwork, signficant time delays, and uncertainty involved in obtaining new visas for technical work in the US for persons outside the NAFTA area. The home page of Genetic Programming Inc. at www.genetic-programming.com and the home page of John R. Koza at Stanford University. koza@genetic-programming.com John Koza Genetic Programming, Inc. P. O. Box K Los Altos, CA 94023-4011 USA ------------------------------ Date: Sat, 24 Jun 2000 10:32:25 +0200 From: Marco DORIGO Subject: Metaheuristics Network: doctoral and postdoctoral positions METAHEURISTICS NETWORK: DOCTORAL AND POSTDOCTORAL POSITIONS AVAILABLE The Metaheuristics Network, a three-year Research Network funded by the European Commission within the Improving Human Potential Programme, will start its activities on September 1st, 2000. The goal of the Metaheuristics Network, which consists of the six laboratories listed below, is to deepen the understanding of metaheuristics so that they can be applied more effectively to the solution of important practical combinatorial optimisation problems. The following doctoral and post-doctoral research positions are expected to become available (subject to funding): IRIDIA, Universite' Libre de Bruxelles, Belgium (1 postdoc, 2 doctoral students) EINDHOVEN University of Technology, The Netherlands (1 doctoral student) EUROBIOS, Paris, France (1 postdoc) IDSIA, Manno, Switzerland (1 postdoc) INTELLEKTIK, Technische Universitaet Darmstadt, Germany (1 postdoc or 2 doctoral students) NAPIER University, Edinburgh, UK (possibly moving to South Bank University, London, UK) (1 doctoral student) The people we are looking for should either possess a PhD (postdoc positions) or a degree that allows them to embark in a doctoral program (doctoral student positions). In both cases, their area of competence should be in at least one of the following disciplines: Computer Science, Operations Research, Computational Intelligence. They should be experienced programmers in procedural or object oriented programming languages and should have knowledge of modern operating systems. The right people will have a commitment to research and publication, and possess good communication and presentation skills. The ability to work as part of a team producing pre-defined deliverables to fixed deadlines is essential. For an informal discussion of these positions and for further details please contact: IRIDIA, Université Libre de Bruxelles: Dr. Marco Dorigo - mdorigo@ulb.ac.be EINDHOVEN University of Technology: Dr. Huub ten Eikelder - wsinhte@win.tue.nl EUROBIOS, Paris: Dr. Eric Bonabeau - Bonabeau@aol.com IDSIA, Manno, Switzerland: Dr. Luca Gambardella - luca@idsia.ch INTELLEKTIK, Technische Universitaet Darmstadt: Dr. Thomas Stuetzle - tom@intellektik.informatik.tu-darmstadt.de NAPIER University - Ben Paechter - benp@dcs.napier.ac.uk Conditions for EU grants The researcher must be 35 years old or less at the time of his appointment (allowances are possible for special cases like child care and military service). The researcher must be a holder of a doctoral degree or of a degree in an appropriate subject in Science or Engineering. The appointment will be for a fixed-term. The applicant must be a national of a Member State of the European Community or of an Associated State or have resided in the European Community for at least five years prior to his appointment. The applicant must choose a Centre located in a state different from his national state and he must not have carried out his activities in that state for more than 12 of the 24 months prior to his appointment. We explicitly encourage female researchers to apply for the offered positions. We guarantee that the selection process, based solely on the research records, will give equal opportunities to female and male researchers. Marco Dorigo Metaheuristics Network Co-ordinator SHORT DESCRIPTION OF THE METAHEURISTIC NETWORK GOALS Many problems of enormous practical importance can be modeled as combinatorial optimisation problems; examples are vehicle routing in distribution management, scheduling of jobs in a production line, generating timetables for scheduling meetings, and many more. Among the most effective methods to solve combinatorial optimisation problems are metaheuristics, that is, sets of concepts that can be used to define approximate algorithms applicable to wide sets of different problems. The OVERALL GOAL of the METAHEURISTICS NETWORK is to co-ordinate European research and postgraduate training in the field of metaheuristics for combinatorial optimization. The main scientific goal of the METAHEURISTICS NETWORK is to deepen our understanding of the working principles of metaheuristics through theoretical and experimental research so that they can be applied more effectively to the solution of important practical combinatorial optimisation problems. The results expected are: (i) a scientific advancement in our understanding of existing metaheuristics and the identification of design principles which make a metaheuristic successful; (ii) a scientific advancement in our understanding of how metaheuristics components, problem characteristics and good performance are related, (iii) the problem-oriented design of new hybrid metaheuristics, and (iv) the definition of a conceptual tool to help engineers choosing the most appropriate metaheuristic for a given problem. The METAHEURISTICS NETWORK will be concerned with the study of a wide variety of metaheuristics and NP-hard problems. The young researchers involved in the network will be exposed to a variety of techniques, as well as to the challenge of applying them to many different situations and problems, both in academic and industrial settings. The research subject is therefore very well suited for training purposes because there are many different techniques involved which require evaluation and experimentation by teams of people with different background and experience and working co-operatively to the solution of a common problem. The main training goals of the network are: (i) to let each young researcher learn about all the tested techniques as well as about a number of important optimisation problems, (ii) to let them learn how to design experiments in a rigorous way, and (iii) to develop their project management skills by means of participation in a strictly co-ordinated team work. --- Marco Dorigo, Ph.D. Chercheur Qualifie' du FNRS IRIDIA CP 194/6 Universite' Libre de Bruxelles Avenue Franklin Roosevelt 50 1050 Bruxelles Belgium mdorigo@ulb.ac.be http://iridia.ulb.ac.be/~mdorigo/ Tel +32-2-6503169 Fax +32-2-6502715 Secretary +32-2-6502729 ------------------------------ Date: Sun, 25 Jun 2000 17:20:32 +0100 From: "Roy, Rajkumar" Subject: WSC5 Virtual Exhibition WSC5 Virtual Exhibition 1st Sept - 1st Nov 2000 Organised by: World Federation on Soft Computing (WFSC) "FREE for Academic and Professional Organisations!" 5th On-line World Conference on Soft Computing in Industrial Applications (WSC5) 4-18 Sept 2000 On the Internet (World-Wide Web) Hosted by: Helsinki University of Technology, Finland Cranfield University, United Kingdom Nagoya University, Japan WSC5 Servers: http://www.hut.fi/Units/PowerElectronics/wsc5/ http://www.cranfield.ac.uk/wsc5/ http://www.bioele.nuee.nagoya-u.ac.jp/wsc5/ WSC5 is the premier on-line event on soft computing and its application in industry. The previous events are: WSC1 (http://www.bioele.nuee.nagoya-u.ac.jp/wsc1/), WSC2 (http://www.bath.ac.uk/Departments/Eng/wsc2/), WSC3 (http://www.cranfield.ac.uk/wsc3/), WSC4 (http://www3.muroran-it.ac.jp/wsc4/index.html) WSC5 will bring together practitioners and researchers in soft computing and its application in industry across the world with the aim to publish quality research rapidly and with less cost. VIRTUAL EXHIBITION: WSC5 conference is also organising a Virtual Exhibition of commercial, industrial and academic stands as before. The exhibition is organised and hosted by Cranfield University. These will present information on commercial products, industrial projects and academic research groups related to soft computing and its industrial application. For more information about the rates for the exhibition etc.: http://www.cranfield.ac.uk/wsc5/vex/vex_info.htm For the exhibition, please provide Dr. Rajkumar Roy, Virtual Exhibition Chair with a small logo (50x50 pixels size, GIF format, no animation please), and a description of your exhibition (max. twenty five words, text format). INDUSTRIAL STAND: Commercial and Industrial organisations can advertise their products at a very nominal cost! You may also win 'The Best Industrial Stand Award'! ACADEMIC STAND: The academics and Professional Institutions can now join the virtual exhibition FREE OF CHARGE. The academic stand would display research activities of different groups in the area, on-line demonstration of products, etc. There is an award for 'The Best Academic Stand' too! SCHEDULE: Deadline for Booking Exhibition Stand: 18 August 2000 WSC5 Virtual Exhibition Opens: 1 Sept 2000 WSC5 Virtual Exhibition Closes: 1 November 2000 For FURTHER INFORMATION on the Virtual Exhibition and for Booking a stand, please contact the Virtual Exhibition Chair: Dr. Rajkumar Roy WSC5 Virtual Exhibition chair Lecturer, Dept. of Enterprise Integration School of Industrial and Manufacturing Science Cranfield University Cranfield, Bedford MK43 0AL United Kingdom Tel. +44 (0)1234 754073 or 750111 Ext. 2423 Fax. +44 (0)1234 750852 Email: r.roy@cranfield.ac.uk or r.roy@ieee.org URL: http://www.cranfield.ac.uk/sims/cim/people/roy.htm ------------------------------ Date: Mon, 26 Jun 2000 14:03:11 +1000 From: Marcus Gallagher Subject: PhD Thesis available: MLP Error Surfaces I am happy to announce the availability of my PhD thesis for download in electronic format. Apologies if you receive multiple copies of this posting. URL: http://www.elec.uq.edu.au/~marcusg/thesis.html Regards, Marcus Multi-Layer Perceptron Error Surfaces: Visualization, Structure and Modelling Marcus R. Gallagher PhD Thesis, University of Queensland, Department of Computer Science and Electrical Engineering, 2000. Abstract The Multi-Layer Perceptron (MLP) is one of the most widely applied and researched Artificial Neural Network model. MLP networks are normally applied to performing supervised learning tasks, which involve iterative training methods to adjust the connection weights within the network. This is commonly formulated as a multivariate non-linear optimization problem over a very high-dimensional space of possible weight configurations. Analogous to the field of mathematical optimization, training an MLP is often described as the search of an error surface for a weight vector which gives the smallest possible error value. Although this presents a useful notion of the training process, there are many problems associated with using the error surface to understand the behaviour of learning algorithms and the properties of MLP mappings themselves. Because of the high-dimensionality of the system, many existing methods of analysis are not well-suited to this problem. Visualizing and describing the error surface are also nontrivial and problematic. These problems are specific to complex systems such as neural networks, which contain large numbers of adjustable parameters, and the investigation of such systems in this way is largely a developing area of research. In this thesis, the concept of the error surface is explored using three related methods. Firstly, Principal Component Analysis (PCA) is proposed as a method for visualizing the learning trajectory followed by an algorithm on the error surface. It is found that PCA provides an effective method for performing such a visualization, as well as providing an indication of the significance of individual weights to the training process. Secondly, sampling methods are used to explore the error surface and to measure certain properties of the error surface, providing the necessary data for an intuitive description of the error surface. A number of practical MLP error surfaces are found to contain a high degree of ultrametric structure, in common with other known configuration spaces of complex systems. Thirdly, a class of global optimization algorithms is also developed, which is focused on the construction and evolution of a model of the error surface (or search space) as an integral part of the optimization process. The relationships between this algorithm class, the Population-Based Incremental Learning algorithm, evolutionary algorithms and cooperative search are discussed. The work provides important practical techniques for exploration of the error surfaces of MLP networks. These techniques can be used to examine the dynamics of different training algorithms, the complexity of MLP mappings and an intuitive description of the nature of the error surface. The configuration spaces of other complex systems are also amenable to many of these techniques. Finally, the algorithmic framework provides a powerful paradigm for visualization of the optimization process and the development of parallel coupled optimization algorithms which apply knowledge of the error surface to solving the optimization problem. Keywords: error surface, neural networks, multi-layer perceptron, global optimization, supervised learning, scientific visualization, ultrametricity, configuration space analysis, search space analysis, evolutionary algorithms, probabilistic modelling, probability density estimation, principal component analysis. ------------------------------ Date: Wed, 28 Jun 2000 12:11:23 +0100 (BST) From: Andrew Tuson Subject: EPSRC PhD Studentship in AI at City University Hi, We currently have 1 (or possibly 2) EPSRC PhD studentships at the School of Informatics, City University, London, UK that we would like to fill ASAP. AI research at City covers a wide range of interests, e.g. - Multi Agent Systems - Neural Networks - Data Mining - Intelligent Optimisation and Evolutionary Computing - Genetic Programming - Logic Programming and Constraints - Bioinformatics - Music and Creativity There is a *possibility* of a CASE award in the area of bioinformatics. **** Those interested please contact me in the first instance. **** The EPSRC eligibility criteria are (in *outline*): - Applicant must be a UK citizen, or have a "relevant connection to the UK" (eg. been granted "indefinite leave to remain in the UK"). - Applicant must have a 2:1 honours degree in a relevant subject; or a 2:2 + MSc; or equivalent. EU students are eligible for a fees-only award (sorry!). *** IMPORTANT - we have no other funding available so non-eligible applicants will be wasting their time *** For the *definitive* guide to (the involved) EPSRC eligibility criteria and terms of the studentship, you are directed to: http://www.epsrc.ac.uk/ Again please contact me if interested. They'll be snapped up soon so I would RSVP ASAP...:-) Andrew Tuson MA(Oxon) MSc(Edin) GRSC - Email: andrewt@soi.city.ac.uk Lecturer, Department of Computing, City University, London, UK. ------------------------------ Date: Thu, 29 Jun 2000 16:42:27 -0400 (EDT) From: spears@AIC.NRL.Navy.Mil (William Spears) Subject: FOGA 2000 Program The FOGA 2000 program is now available at: http://www.cs.virginia.edu/~wnm/FOGA2000/program.html Just a reminder - those who wish to register should do so as soon as possible, to ensure a room in the hotel (there are rooms still available). Cheers, Bill ------------------------------ Date: Fri, 30 Jun 2000 13:56:59 +0100 From: miguel.goulao@di.fct.unl.pt Subject: CSMR 2001 call for papers Fifth European Conference on Software Maintenance and Reengineering Lisbon area, Portugal 14 to 16 March 2001 http://www.esw.inesc.pt/csmr2001 Call for Papers CSMR is the premier European Conference on Software Maintenance and Reengineering. Its purpose is to promote both discussion and interaction about maintenance and reengineering. Topics of interest include, but are not restricted to: * Maintenance and reengineering metrics and economics * Patterns languages for maintenance and reengineering * Experience reports on maintenance and reengineering * Maintenance and reengineering tools * Enabling technologies for maintenance and reengineering * Formal methods to support maintenance and reengineering * Software evolution and architecture recovery * System assessment for reengineering or maintenance * Migration and maintenance issues * Dealing with legacy systems towards new technologies One of the basic intentions of this conference is to offer an European forum for discussion and exchange of experiences among researchers and practitioners. Therefore, besides academics, we kindly invite all those in companies developing maintenance tools, offering reengineering services or going through legacy systems migration experiences to contribute by submitting papers or presenting innovative tools, solutions or experience reports. This conference is not limited to European participants. Authors from outside Europe are especially welcomed. SUBMISSIONS: Two types of submissions will be accepted: full length papers (not exceeding 4000 words in length and including a 150-200 word abstract) and short papers (not exceeding 2000 words in length and including a 75-100 word abstract). All papers should be in English. Authors are requested to submit electronically a PostScript or PDF version of their papers. In addition, they should send a separate file containing the title of the paper, full names, affiliations, postal and e-mail addresses, fax and telephone numbers of all authors. We encourage authors to make submissions through the web based submission system that will be available. For submission details please look at the conference web site. IEEE Computer Society Press will publish the CSMR=922001 Proceedings. Full papers exceeding 10 pages (4 pages for short papers) will be charged for pages in excess. Authors of accepted papers must sign the IEEE copyright form. At least one author of each accepted submission should register and present the paper at the conference. The official language will be English. IMPORTANT DATES: DEADLINE for submissions - Friday, September 15, 2000 Author's notification - Monday, November 20, 2000 DEADLINE for camera-ready of accepted papers - Friday, December 15, 2000 SPECIAL SESSIONS: Sessions of special interest proposed by delegates will be welcomed. Please send suggestions to the program chair before the submissions closing date. -- Miguel Goulão Departamento de Informática - FCT/UNL email: mg@di.fct.unl.pt Tel: 351 21 2948536 Ext: 0749 (DI) 351 21 2948300 Ext: 0749 (Geral) [ This submission has been shortened. For more information, see http://www.esw.inesc.pt/csmr2001 -Moderator ] ------------------------------ Date: Tue, 4 Jul 2000 19:10:20 +0100 From: "Taylor, Tim" Subject: MathEngine Evolved Creatures Dear colleagues, I am pleased to say that I have, at last, prepared a webpage describing the work that Colm Massey and I did at MathEngine last year on re-implementing Karl Sims' evolved creatures work. I was hoping to get the demos in a state such that they could be run in realtime in MathEngine over the web. However, this would have required more time than I had to spend, so I have instead produced mpeg movies of some example creatures. The mpeg encodings aren't great, but at least you can get some idea of what the creatures are doing... You can find the webpage at: http://computing.tay.ac.uk/timtaylor/demos/mathengine/ Enjoy! Tim --- Tim Taylor Research Associate, School of Computing, University of Abertay Dundee Kydd Building, Bell Street, Dundee DD1 1HG, Scotland tel. +44-(0)1382-308959 fax. +44-(0)1382-308627 mailto:tim.taylor@abertay.ac.uk http://computing.tay.ac.uk/timtaylor/ ------------------------------ Date: Wed, 05 Jul 2000 14:21:01 +0900 From: "Uwe R. Zimmer" Subject: Deadline extention - ISI'2001 workshop !!! DEADLINE FOR SUBMISSIONS EXTENDED: JULY 31. !!! So, there is still a chance to submit a manuscript (extended abstract or full paper) to the ... Workshop on: Autonomous artificial systems exploring hostile environments http://www.gmd.gr.jp/JRL/events.html as part of the International NAISO Congress on Information Science Innovations (ISI'2001) March 18th, 2001 at the American University of Dubai, U.A.E. [ This submission has been shortened. For more information, see http://www.gmd.gr.jp/JRL/events.html -Moderator ] ------------------------------ Date: Wed, 5 Jul 2000 18:47:55 +0100 (BST) From: Xin Yao Subject: JOB: Research Fellow, The University of Birmingham, England Hi, The deadline for the following position has been extended to 27/7/2000. (The previous deadline was only one day after it appeared in GA-List.) I'm happy to talk to potential candidates at the upcoming CEC'2000 in San Diego. Best regards, Xin RESEARCH FELLOW IN EVOLUTIONARY COMPUTATION (Ref. No. S35236/00) http://www.bham.ac.uk/personnel/s35236.htm Applications are invited for a research fellowship in evolutionary computation (available immediately for up to three years, full-time) in the School of Computer Science, the University of Birmingham, Edgbaston, Birmingham, B15 2TT, England. We are particularly interested in candidates with a background in co-evolution, evolvable hardware or a closely related topic. However, outstanding applicants from any areas of evolutionary computation will be considered seriously. Applicants should have or be about to complete a PhD in Computer Science, Computer Engineering, Electrical Engineering, or a closely related field. The successful candidate is expected to have research experience and record of outstanding quality in evolutionary computation or a closely related area, as evidenced by publications in leading international journals or conference proceedings. The research potential of a new PhD (or a nearly completed PhD) may also be judged from his/her PhD thesis. Strong background and experience in computational studies and excellent analytical and programming skills will be highly valued. The successful applicant must be able to work well in a team environment. The School has a very strong group in evolutionary computation with international reputation. The current research interests in the school include, but are not limited to, co-evolution, niching and speciation, evolutionary games, evolvable hardware, evolutionary artificial neural networks, genetic programming, knowledge transfer in evolutionary learning systems, evolvable architectures for human-like minds, data mining, complexity analysis of evolutionary algorithms, schema theorems, global optimisation and constrained optimisation. The successful applicant is required to contribute to the School's teaching activities, including teaching one module per year and helping occasionally with another. The areas of teaching will be in neural networks and evolutionary computation. The starting salary for the post is GBP16,286 - GBP24,479 per annum (Depending on experience and qualifications). For further particulars, please visit http://www.bham.ac.uk/personnel/s35236.htm For informal enquiries, please contact Prof Xin Yao, phone (+44) 121 414 3747, email: X.Yao@cs.bham.ac.uk. CLOSING DATE FOR RECEIPT OF APPLICATIONS: 27 July 2000 (late application may be considered) APPLICATION FORMS RETURNABLE TO The Director of Personnel Services The University of Birmingham Edgbaston, Birmingham, B15 2TT England RECRUITMENT OFFICE FAX NUMBER +44 121 414 4802 RECRUITMENT OFFICE TELEPHONE NUMBER +44 121 414 6486 RECRUITMENT OFFICE E-MAIL ADDRESS h.h.luong@bham.ac.uk ------------------------------ Date: Thu, 6 Jul 2000 10:10:41 -0700 From: "John Koza" Subject: PhD Thesis Research / Post-Doc Fellowship for Research in GP 3-YEAR RESEARCH GRANT IN FOR PhD THESIS RESEARCH OR POST-DOCTORAL RESEARCH IN APPLYING GENETIC PROGRAMMING TO REAL-WORLD IMAGE ANALYSIS, OBJECT DETECTION, AND PATTERN RECOGNITION DERA is seeking applicants for a 3-year research grant for PhD thesis research or post doctoral research in applying genetic programming to the fields of real-world image analysis, object detection, and pattern recognition. The grant is available to either (1) a student who has passed all applicable department examinations and requirements to undertake PhD thesis research or (2) a post-doctoral researcher. The grant may be used at any university worldwide. The grant will provide a monthly stipend competitive with those provided at the university involved to PhD students or post-doctoral researchers, tuition (in the case of students) or a research grant to the university involved (in the case of postdoctoral researcher), an allowance for computer equipment, and a travel allowance for attending scientific conference(s) and visiting DERA in Malvern, England at least once a year to consult on research progress and present results. The grant will be renewed annually provided the recipient is making satisfactory progress in his or her research. The Defence Evaluation and Research Agency (DERA) is an Agency of the Ministry of Defence (MOD) of the United Kingdom and incorporates the bulk of the MOD's non-nuclear research, technology and test and evaluation establishments. DERA (http://www.dera.gov.uk) is one of Europe's largest single research organizations. Among its 12,000 staff DERA employs many leading scientists and internationally acclaimed experts. It offers a unique range of services, from the highest level of operational studies and analysis, through the various categories of basic and applied research, to consultancy advice on the procurement process and the test and evaluation of specific equipment in both the development phase and during actual operations. DERA's test facilities include indoor and outdoor ranges for weapon effectiveness trials, underwater target ranges and marine testing facilities, automotive test tracks, wind tunnels and climatic testing laboratories. DERA's pioneering research and development this century includes the invention of liquid crystal displays (LCDs), carbon fibre, the technology for flat panel speakers, infra-red sensors and microwave radar, as well as Chobham armor, and shaped charges. DERA has combined and inherited the technology and expertise from Government defence research sites throughout the United Kingdom. PRELIMINARY INTERVIEWS AT GECCO-2000, EH-2000, AND CEC-2000 CONFERENCES Applications received by August 15, 2000 will be promptly reviewed. If a suitable candidate is identified, an award will be made with a starting date in the early fall 2000. If no suitable candidate has applied by this first date, the awards committee will continue to receive applications until November 1, 2000. Then, if a suitable candidate is identified, an award will be made with a starting date in early 2001. The application form and other details will be posted at http://apps.dera.gov.uk/job_search/jobsrtn.asp no later than July 26, 2000. Potential applicants will find it advantageous to make arrangement to informally meet with DR. DANIEL HOWARD, chair of the grant committee, during the GECCO -2000, EH-2000, or CEC-2000 conferences to briefly discuss their research interests and qualifications. One or more additional members of the awards committee may also be present at these conferences for these informal interviews. A preliminary interview is highly desirable, but not required in order to apply. Dr. Howard will be staying at the Riviera Hotel for the GECCO-2000 conference in Las Vegas from July 9 through July 12 (Phone: 702-734-5110 ), at the Crowne Plaza Cabana Hotel for the EH-2000 conference in Palo Alto from July 13 to 15 (Phone: 650-8570787), and at the San Diego Marriott La Jolla Hotel for the CEC-2000 conference from July 16 to 20 (Phone: 858-587-1414). Potential candidates may contact Dr. Howard to set up an appointment during one of the 3 conferences. ------------------------------ Date: Wed, 12 Jul 2000 13:42:32 +0200 (MET DST) From: Thomas Stuetzle Subject: PhD positions at TU Darmstadt, Germany Metaheuristics Network PhD and PostDoc Positions available at Network Node TU Darmstadt Many problems of enormous practical importance can be modeled as combinatorial optimization problems; examples are vehicle routing in distribution management, scheduling of jobs in a production line, generating timetables for scheduling meetings, and many more. Among the most effective methods to solve combinatorial optimization problems are metaheuristics, that is, sets of concepts that can be used to define approximate algorithms applicable to wide sets of different problems. The Metaheuristics Network, a Research Network funded by the European Commission within the Improving Human Potential Programme, is concerned with the study of a wide variety of metaheuristics and NP-hard problems. The main scientific goal is to deepen our understanding of the working principles of metaheuristics through theoretical and experimental research so that they can be applied more effectively to the solution of important practical combinatorial optimization problems. The young researchers involved in the network will be exposed to a variety of techniques, as well as to the challenge of applying these techniques to many different situations and problems, both in academic and industrial settings. The research subject is very well suited for training purposes because there are many different techniques involved which require evaluation and experimentation by teams of people with different background and experience and working co-operatively to the solution of a common problem. The Intellectics Group, the Metaheuristics Network node at the computer science department of TU Darmstadt, has very active research lines on local search algorithms and metaheuristics as well as on more traditional Artificial Intelligence techniques. We offer two doctoral and one PostDoc research positions over a three years period starting September 1st 2000. The people we are looking for should possess a PhD or a degree that allows them to embark in a doctoral program. Their area of competence and/or interest should be in at least one of the following disciplines: Computer Science, Operations Research, or Computational Intelligence. They should be experienced programmers in procedural or object oriented programming languages and should have knowledge of modern operating systems. The working language will be English. The applicants will have a commitment to research and publication, and possess good communication and presentation skills (in English). They will be available for traveling between European labs participating in the research project, located in Belgium, France, Germany, The Netherlands, United Kingdom, and Switzerland. Application deadline There is no strict submission deadline; positions will be filled as adequate candidates become available. We suggest you contact the local network co-ordinator as soon as possible. Contact information If you wish to apply, please contact Thomas Stuetzle (stuetzle@informatik.tu-darmstadt.de, +49 6151 166651) or send an application (in English, including detailed CV and references) to Thomas Stuetzle, TU Darmstadt, CS Department, Intellectics Group, Alexanderstr. 10, 64283 Darmstadt, Germany. Fellowship amount The amount of the fellowship for PhD students is DM 2690.- net and for PostDocs approx. DM 2900.- net. Conditions for EU grants The most important conditions for EU grants are that (i) the researcher must be 35 years old or less at the time of his appointment (allowances are possible), (ii) the applicant must be a national of an EU member state or of an Associated State or have resided in the EU for at least five years prior to his appointment, and (iii) fulfill certain mobility requirements. For more details see the WWW-page http://www.intellektik.informatik.tu-darmstadt.de/~tom/TUD-Openings.html --- Dr. Thomas Stuetzle Darmstadt University of Technology (TUD) Computer Science Department Alexanderstr. 10 64283 Darmstadt Tel.: +49 6151 16-6651 Fax.: +49 6151 16-5326 Email: stuetzle@informatik.tu-darmstadt.de WWW: http://www.intellektik.informatik.tu-darmstadt.de/~tom ------------------------------ Date: Mon, 17 Jul 2000 18:38:51 +0200 (MET DST) From: Evelyne.Lutton@inria.fr Subject: PPSN 2000 - Paris, France, 16-20 Sept. 2000 PPSN VI CALL FOR PARTICIPATION THE SIXTH INTERNATIONAL CONFERENCE on PARALLEL PROBLEM SOLVING FROM NATURE PPSN VI Paris, France, September 16 - 20, 2000 DEADLINE FOR EARLY REGISTRATION : August 25, 2000. http://www.inria.fr/PPSN2000/ IMPORTANT INFORMATION FOR STUDENTS : The International Society on Genetic and Evolutionary Computation (ISGEC) is kindly offering some student travel grants for attending PPSN. Applicants should send a mail stating their motivation, a small resume in Postscript and a proof of their student status by mail to PPSN organizing Committee (PPSN2000@inria.fr). The grants will be allocated by the Organizing Committee with priority to authors of accepted papers, and based on the cost of travel from the student's home to Paris. The exact amount for each grant will depend on the number of relevant applications. DEADLINE FOR APPLICATION IS JULY 23. All applicants will be personally notified by July 31. ------------------------------ Date: Tue, 18 Jul 2000 13:54:55 +0200 (MET DST) From: Yaochu Jin Subject: Errata on "Evolutionary optimization ..." in GECCO-2000 Proc. Dear all, Due to unknown reasons, all dotted and dashed lines are missing from the figures(Fig.3-Fig.9) in the paper "On evolutionary optimization with approximate fitness functions" in GECCO-2000 Proceedings, pp.786-793. A prelinminary version of this paper with correct figures can be downloaded from the following website: http://www.soft-computing.de/ I apologize for any inconvenience. Best regards Yaochu Jin --- Yaochu Jin Future Technology Research Honda R&D Europe (D) GmbH Carl-Legien-Strasse 30 63073 Offenbach/Main Germany Tel: +49 69 89011735 Fax: +49 69 89011749 Email: yaochu.jin@hre-ftr.f.rd.honda.jp Alias: yaochu.jin@ieee.org http://www.soft-computing.de/ ------------------------------ Date: Tue, 18 Jul 2000 08:14:25 -0700 (PDT) From: Hillol Kargupta Subject: Special Issue of the Complex Systems Journal SPECIAL ISSUE ON THE COMPUTATION IN GENE EXPRESSION Complex Systems Journal Guest Editors: Hillol Kargupta Washington State University, USA hillol@eecs.wsu.edu Dirk Thierens, Universiteit Utrecht, The Netherlands Dirk.Thierens@cs.uu.nl Web site: http://www.eecs.wsu.edu/~hillol/GECCO/gene-expression-complex-systems.html Overall Description: The evaluation of the DNA through the construction of mRNA and protein is called gene expression. Since the structure of the protein determines the efficacy of the DNA, gene expression can be viewed as the process of evaluating the evolutionary fitness function. For some reason nature seems to transform the genetic representation of the DNA to mRNA and subsequently to protein before evaluating its fitness. Representation plays an important role in problem solving which is widely acknowledged in many fields such as physics, mathematics, engineering, machine learning, optimization and many others. Representation transformations are often used in these fields for solving problems efficiently. Therefore representation transformations in gene expression allude intriguing possibilities. However, very little is understood about the role of gene expression in evolutionary search. This special issue will focus on exploring gene expression based on our basic understanding of evolutionary search, learning, and optimization. The topics of interest include, but are not limited to: 1) Theoretical and experimental analysis of gene expression operators. 2) Relation of gene expression and efficient, scalable genetic and evolutionary computation. 3) Evolutionary algorithms motivated by gene expression. 4) Insightful applications of gene expression-based computation. Deadlines Paper submission October 16, 2000 Acceptance notice November 30, 2000 Final version due January 10, 2001 Submission Instruction Authors may submit their papers (in postscript, MS word or PDF version through e-mail) to the following address. Hillol Kargupta Faculty of Computer Science School of Electrical Engineering and Computer Science Washington State University Pullman, WA 99164-2752 (509) 335-6602 (509) 338-3818 (Fax) hillol@eecs.wsu.edu Final version of the paper will be required to be in the format prescribed by the Complex Systems Journal. For details please see http://www.complex-systems.com ------------------------------ Date: Sat, 22 Jul 100 13:41:40 -0600 (MDT) From: Xindong Wu Subject: Knowledge and Information Systems: Vol 2 No 3 (2000) Knowledge and Information Systems: An International Journal ISSN 0219-1377 by Springer-Verlag Home Page: http://kais.mines.edu/~kais/ Volume 2, Number 3 (August 2000): Table of Contents Critical Reviews - The State of the Art in Agent Communication Languages by Mamadou Tadiou Kone, Akira Shimazu and Tatsuo Nakajima Regular Papers - Simulating the Ecology of Oligopoly Competition with Genetic Algorithms by Shu-Heng Chen and Chih-Chi Ni - Intentions in the Coordinated Generation of Graphics and Text from Tabular Data by Massimo Fasciano and Guy Lapalme - An Intelligent Decision Support System for Investment Analysis by K.L. Poh Short Papers - Spatio-Temporal Analysis with the Self-Organizing Feature Map by Susan E. George - A Probe-based Technique to Optimize Join Queries in Distributed Internet Databases by Cyrus Shahabi, Latifur Khan, and Dennis McLeod ------------------------------ Date: Sun, 23 Jul 2000 16:07:17 -0000 From: "Candida Ferreira" Subject: Gene expression programming: a new GP-style technique Hi, I invented a technique inspired in gene expression and I made a site to divulge it and hopefully find someone interested in sponsoring the project. Please take a look! www.gene-expression-programming.com All feedback will be greatly appreciated. See you there! Candida Ferreira ------------------------------ Date: Mon, 24 Jul 2000 13:43:00 +0100 From: j.n.carter@ic.ac.uk Subject: Vacant Post: Imperial College of Sci Tech and Medicine Imperial College of Science Technology and Medicine T.H. Huxley School of Environment, Earth Sciences and Engineering Computational Physics and Geophysics Post-Doctorial Research Assistant in Evolutionary Computation Applications are invited for a research assistant in Evolutionary Computation, for up to 39 months, in the Computational Physics and Geophysics Research Group, to work with Dr Jonathan Carter on the use of Genetic Algorithms to improve oil reservoir characterisation. Applicants will ideally have, or will be about to complete, a PhD in a numerical area of science or engineering, or will have relevant experience. The sucessful candidate is likely to have research experience in evolutionary computation and/or numerical computing/programming. Reservoir characterisation aims to help maximise the ammount of oil/gas recovered from a petroleum reservoir - a vital issue in sustainable development. The research will focus on automating the definition of the numerical model (ie estimating the model parameters that give the best fit to the measured data) and will make innovative use of Genetic Algorithms so as to minimise the number of models that need to be tested. There should be opportunities to test the new method on real world problems in collaboration with commercial companies. Salary range will be on the RAIA scale (£20,865 to £27,347 inclusive of London Allowance). Applications, including a CV, list of publications and the names of two referees (who should be asked to send their references to the same address), should be sent to Dr J.N. Carter, Computational Physics and Geophysics Research Group, T.H. Huxley School of Environment, Earth Sciences and Engineering, Prince Consort Road, South Kensington, London, UK. Applications may also be emailed (as plain ascii text, postscript or pdf files - MSWord documents will not be accepted) to "j.n.carter@ic.ac.uk". Closing date: Sunday 20th August 2000 The College is striving towards Equal Opportunities This advert also appear at http://www.jobs.ac.uk ------------------------------ Date: Mon, 24 Jul 2000 09:09:16 -0400 From: Stewart Wilson Subject: New Classifier Systems Book LEARNING CLASSIFIER SYSTEMS: FROM FOUNDATIONS TO APPLICATIONS has just been published by Springer. Please see the announcement at http://www.springer.de/cgi-bin/search_book.pl?isbn=3-540-67729-1 You can download the Table of Contents and Preface. Edited by P. L. Lanzi, W. Stolzmann, and S. W. Wilson, the book provides a unique survey of the current state of the art of LCS and highlights some of the most promising research directions. In the first part, eleven well-known researchers offer answers to the question, "What is a learning classifier system?" The next three articles offer a "roadmap" to the past decade of LCS research, a description of the state of XCS research, and an introduction to fuzzy LCS's. The second part has articles on advanced topics of current interest, including alternative representations, methods for evaluating rule utility, and extensions to existing classifier system models. The third part contains articles on promising applications in areas like data mining, medical data analysis, economic trading agents, aircraft maneuvering, and autonomous robotics. An appendix comprising 467 entries provides a comprehensive LCS bibliography. 347 pp., DM 86 (approximately 43 USD) ------------------------------ End of Genetic Algorithms Digest ******************************