GECCO 2010

1

Black Box Optimization Benchmarking 2010 (BBOB 2010)

In the last decades, numerous algorithms taking inspiration from nature have been proposed to handle continuous optimization problems: Evolution Strategies and Evolutionary Programming, real-coded Genetic Algorithms, Estimation of Distribution Algorithms, Differential Evolution, Particle Swarm Optimization, memetic algorithms, to name just a few. On the other hand, gradient-based, quasi-Newton algorithms, pattern search algorithms and more recently Derivative-Free-Optimization algorithms have been proposed and improved by applied mathematicians. Although comparisons of the performance of different optimizers are made in research studies, few optimizers are usually tested within a single work. Different works usually do not use the same experimental settings hampering their comparability, and often comparisons are biased to certain types of test functions: for a long time, most Evolutionary Algorithms have been tested mainly on separable problems.

The BBOB workshop pursues efforts for a comprehensive comparison of continuous search algorithms for single-objective black-box optimization. As a follow-up of BBOB at GECCO 2009, the workshop exploits the previously developed framework allowing to rigorously compare continuous optimizers for non-noisy and noisy optimization. Apart from minor adjustments, we provide the same testbed and experimental setting as last year. Participants will have access to data that have been submitted for BBOB 2009 and are provided with Python scripts to easily compare the performance of any two algorithms in detail. Participants are invited to submit a paper with the results of their algorithm of choice, and/or comparison with an algorithm from our database, and/or comparison of two or several algorithms. An overall analysis and comparison will be accomplished by the organizers and presented during the workshop together with the single presentations of each participant. A plenary discussion on future improvements will, among others, address the question, of how the testbed should evolve.

Visit this website for more information and code downloads.

Any further questions? mailto bboblri.fr

Anne Auger

Dr. Anne Auger received her diploma in mathematics from the University of Paris VI, France, in 2001. She also obtained the french highest diploma for teaching mathematics, "Agregation de mathematiques". She received the doctoral degree from the university Paris VI in 2004. Afterwards, she worked for two years (2004-2006) as a postdoctoral researcher at ETH (in Zurich) in the Computational Laboratory (CoLab). Since October 2006, she holds a permanent research position at INRIA (French National Research Institute in Computer Science and Applied Mathematics). Her research interests are stochastic continuous optimization, including theoretical analyses of randomized search heuristics. She published more than fifteen articles at top conferences and journals in this area. She organized (and will organize) the biannual Dagstuhl seminar "Theory of Evolutionary Computation" in 2008 (and 2010).

Hans-Georg Beyer

He received the Diploma degree in Theoretical Electrical Engineering from the Ilmenau Technical University, Germany, in 1982 and the Ph.D. in physics from the Bauhaus-University Weimar, Germany, in 1989. He finished his Habilitation thesis at the University of Dortmund, Germany, and qualified for a professorship in Computer Science in 1997.

From 1982 to 1984, he worked as an R&D Engineer in the Reliability Physics Department, VEB Gleichrichterwerk, Stahnsdorf, Germany. From 1984 to 1989, he was Research and Teaching Assistant and later on Postdoc at the Physics Department and the Computer Science Department, Bauhaus-University Weimar. From 1990 to 1992, he worked as a Senior Researcher in the Electromagnetic Fields Theory Group at the Darmstadt University of Technology, Germany. From 1993 to 2004 he was with the Computer Science Department of the University of Dortmund. In 1997 he became a DFG (German Research Foundation) Heisenberg Fellow. He was leader of a working group and from 2003 to 2004 professor of computer science. Since 2004 he is professor of computer science at the Vorarlberg University of Applied Sciences, Austria. His main research area is the theory of evolutionary algorithms. He is author of the book "The Theory of Evolution Strategies" (Heidelberg: Springer-Verlag, 2001) and author/coauthor of more than 100 papers.

Dr. Beyer serves as an Associate Editor for the IEEE Transactions on Evolutionary Computation since 1997 and for the Journal of Evolutionary Computation (MIT-Press). He served as guest editor for the Journals "Natural Computing" and "Genetic Programming and Evolvable Machines." He was co-organizer of the Dagstuhl-Seminar series on ``Theory of Evolutionary Algorithms'' (in 2000, 2002, and 2004). He served as Editor-in-Chief of the Genetic and Evolutionary Computation Conference (GECCO) 2005 and was/is program co-chair of the Conferences GECCO-2000, GECCO-2003, GECCO-2004, GECCO-2007, GECCO-2009, and Parallel Problem Solving from Nature (PPSN) in 2002 and 2006.

Nikolaus Hansen

He is a research scientist at The French National Institute for Research in Computer Science and Control (INRIA). Educated in medicine and mathematics, he received a Ph.D. in civil engineering in 1998 from the Technical University Berlin under Ingo Rechenberg. Before he joined INRIA, he has been working in applied artificial intelligence and genomics, and he has done research in evolutionary computation and computational science at the Technical University Berlin and the ETH Zurich. His main research interests are learning and adaptation in evolutionary computation and the development of algorithms applicable in practice. The latter is also the motivation behind his interest and initiative in the benchmarking of algorithms.

Steffen Finck

He received the Diploma in aeronautical engineering from the University of Stuttgart, Stuttgart, Germany in 2006. During his study he participated in an exchange program at the Rose-Hulman Institute of Technology, Terre-Haute, USA from which he received a MSc degree in mechanical engineering in 2004. Since end of 2006 he is a PhD-student at the FH Vorarlberg University of Applied Sciences in Dornbirn, Austria. His PhD is concerned with direct search methods under the influence of noise.

Raymond Ros

He graduated from the Ecole Superieure de Physique et de Chimie Industrielles (Paris, France) and obtained a MSc in Computer Science from the Universite Paris-Sud in 2005. Since 2009, he holds a position of expert engineer in the French National Institute for Research in Computer Science and Control (INRIA) as he prepares his PhD under Nikolaus Hansen and Michele Sebag in the University of Paris-Sud on benchmarking continuous optimization algorithms and developing evolutionary algorithms for large-scale problems.

Petr Posik

He recieved his Diploma degree in Technical Cybernetics in 2001 and Ph.D. in Artificial Intelligence and Biocybernetics in 2007, both from the Czech Technical University in Prague, Czech Republic. From 2001 to 2004 he also worked as statistician, analyst and lecturer for StatSoft, Czech Republic. Since 2005 he works as a researcher in the Intelligent Data Analysis Group, Department of Cybernetics at the Czech Technical University. Being on the boundary of optimization, statistics and machine learning, his research interests are aimed at improving the characteristics of evolutionary algorithms with techniques of statistical machine learning. He also serves as a reviewer for several journals and conferences in the evolutionary-computation field.

2

Evolutionary Computation and Multi-Agent Systems and Simulation (ECoMASS) - Fourth Annual Workshop

Evolutionary computation (EC) and multi-agent systems and simulation (MASS) both involve populations of agents. EC is a learning technique by which a population of individual agents adapts according to the selection pressures exerted by an environment; MASS seeks to understand how to coordinate the actions of a population of (possibly selfish) autonomous agents that share an environment so that some outcome is achieved.

Both EC and MASS have top-down and bottom-up features. For example, some aspects of multi-agent system engineering (e.g., mechanism design) are concerned with how top-down structure can constrain individual decisions. Similarly, most work in EC is concerned with how to engineer selective pressures to drive the evolution of individuals towards some desired goal. Multi-agent simulation (also called agent-based modeling) addresses the bottom-up issue of how collective behavior emerges from individual action. Likewise, the study of evolutionary dynamics within EC often considers how population-level phenomena emerge from individual-level interactions. Thus, at a high level, we may view EC and MASS as examining analogous processes. It is therefore natural to consider how EC may be relevant to MASS, and vice versa; indeed, applications and techniques from one field have often made use of technologies and algorithms from the other field.

Example Topics:

Multi-agent systems and agent-based models utilizing evolutionary computation
Optimization of multi-agent systems and agent-based models using evolutionary computation
Evolutionary computation models which rely not on explicit fitness functions but rather implicit fitness functions defined by the relationship to other individuals / agents
Applications utilizing MASS and EC in combination
Biological agent-based models (usually called individual-based models) involving evolution
Evolution of cooperation and altruism
Genotypic representation of the complex phenotypic strategies of MASS
Evolutionary learning within MASS (including Baldwinian learning and phenotypic plasticity)
Emergence and feedbacks
Open-ended strategy spaces and evolution
Adaptive individuals within evolving populations

Workshop website

William Rand

Bill is an assistant professor of Marketing, Decision, Operations, and Information Technology, and Computer Science at the University of Maryland, where he also serves as the Director of Research for the Center for Complexity in Business. He obtained his Ph.D. from the University of Michigan working under Rick Riolo and John Holland. His dissertation focused on the use of GAs in dynamic environments, and at the same time he helped develop a large-scale model of residential land-use decisions in Southeastern Michigan to model the effects of suburban sprawl. Bill is co-authoring a textbook with Uri Wilensky at Northwestern University that is the a hands-on introduction to agent-based modeling.

Rick Riolo

Rick is an Associate Research Scientist and Director of the Computer Lab in the Center for the Study of Complex Systems at the University of Michigan. His research interests include evolutionary algorithms in theory and in models of complex adaptive systems, and agent-based modeling approaches to studying problems across a wide variety of complex systems, e.g., the spread of antibiotic resistance, the effects of phenotypic plasticity on ecological community dynamics, urban sprawl, and the effects of formal and informal institutions on the sustainability of common resource pools.

3

Evolutionary Computation Techniques for Constraint Handling

Evolutionary algorithms are often applied to difficult optimization problems where candidate solutions are subject to various constraints. There is a large body of practical guidelines and theoretical results about evolutionary algorithms in the context of constrained optimization. Closely related is the subject of constraint programming where some solution satisfying all constraints is sought. In both contexts evolutionary computation techniques can be and are applied to solve hard problems. The workshop intends to provide a forum for current research in these areas. Algorithmic techniques, case studies, theoretical and empirical analyses are all welcome. While focusing on handling constraints we intend to provide a forum for discussion, therefore allowing generous time slots for each presentation and ample time for discussions.

More information can be found at the following web page .

Carlos Coello

Carlos Artemio Coello Coello received a BSc in Civil Engineering from the Universidad Autonoma de Chiapas in Mexico in 1991 (graduating Summa Cum Laude). Then, he was awarded a scholarship from the Mexican government to pursue graduate studies in Computer Science at Tulane University (in the USA). He received a MSc and a PhD in Computer Science in 1993 and 1996, respectively. Dr. Coello has been a Senior Research Fellow in the Plymouth Engineering Design Centre (in England) and a Visiting Professor at DePauw University (in the USA). He is currently full professor at CINVESTAV-IPN in Mexico City, Mexico.

He has published over 200 papers in international peer-reviewed journals, book chapters, and conferences. He has also co-authored the book "Evolutionary Algorithms for Solving Multi-Objective Problems", which is now in its Second Edition (Springer, 2007) and has co-edited the book "Applications of Multi-Objective Evolutionary Algorithms" (World Scientific, 2004). He has delivered invited talks, keynote speeches and tutorials at international conferences held in Spain, USA, Canada, Switzerland, UK, Chile, Colombia, Brazil, Argentina, India, Uruguay and Mexico. Dr. Coello has served as a technical reviewer for over 50 international journals and for more than 60 international conferences and actually serves as associate editor of the journals "IEEE Transactions on Evolutionary Computation", "Evolutionary Computation", "Computational Optimization and Applications", "Pattern Analysis and Applications" and the "Journal of Heuristics". He is also a member of the editorial boards of the journals "Soft Computing", "Engineering Optimization", the "Journal of Memetic Computing" and the "International Journal of Computational Intelligence Research". He received the 2007 National Research Award (granted by the Mexican Academy of Science) in the area of "exact sciences". He is member of the Mexican Academy of Science, Senior Member of the IEEE, and member of Sigma Xi, The Scientific Research Society. His current research interests are: evolutionary multiobjective optimization, constraint-handling techniques for evolutionary algorithms and evolvable hardware.

Dara Curran

He received a BSc. in Information Technology from the National University of Ireland, Galway in 2000. After spending some time in industry, he was awarded an Irish Research Council for Research Engineering and Technology scholarship to pursue a PhD, which he obtained from the National University of Ireland, Galway in 2006. He is currently a staff scientist at the Cork Constraint Computation Centre. Prior to this, he was an Enterprise Ireland Principal Investigator and an Irish Research Council for Research, Engineering and Technology Post-doctoral fellow. He has published over 35 research papers and co-organised the first workshop on Organisation, Cooperation and Emergence in Social Learning Agents at ECAL 2009. His research is focused on the application of learning models to evolutionary algorithms and the use of evolutionary computation for Artificial Life simulation.

Thomas Jansen

He was born in 1969, and studied Computer Science at the University of Dortmund, Germany. He received his diploma (1996, summa cum laude) and Ph.D. (2000, summa cum laude) there. From September 2001 to August 2002 he stayed as a Post-Doc at Kenneth De Jong's EClab at the George Mason University in Fairfax, VA. He was Juniorprofessor for Computational Intelligence from September 2002 to February 2009 at the Technical University Dortmund. Since March 2009 he is Stokes Lecturer at the Department of Computer Science at the University College Cork, Ireland. His research is centered around design and theoretical analysis of evolutionary algorithms and other randomized search heuristics. He has published 14 journal papers, 29 conference papers and contributed six book chapters. He is associate editor of Evolutionary Computation (MIT Press), member of the steering committee of the Theory of Randomized Search Heuristics workshop series, was program chair at PPSN 2008, co-organized FOGA 2009, a workshop on Bridging Theory and Practice (PPSN 2008) and two Dagstuhl workshops on Theory of Evolutionary Computation (2004 and 2006).

4

Entropy, Information and Complexity

Among the basic concepts and state variables of thermodynamics, ENTROPY plays a central role, explained by Boltzmann et al as statistical mechanics. Entropy and its complement [mutual] INFORMATION play key roles in the theory of communication by Shannon et al. COMPLEXITY and algorithmic information play the corresponding roles in the complexity theory of Kolmogorov et al.

All 3 of these theories are directly applicable to both natural biological evolution and evolutionary computation, but have generally been applied by neither theoreticians nor practitioners in the latter field. In recent years, a few papers have appeared, scattered across different venues. This workshop will bring together researchers with a common interest in applying these concepts systematically to evolutionary computation and learning.

The first half of the workshop will be a series of mini-tutorials on the related concepts and variables in the context of each of the 3 theories.

The second half of the workshop will be a series of presentations of papers describing original work using these concepts in evolutionary algorithm theory, design and application.

A formal, unified, information theoretic and thermodynamic framework of evolutionary computation and learning would be of immense value to theorists and practitioners, so prospects for this, and any other topics of interest to the participants, will be brainstormed.

Workshop website

Stu Card

He is Chief Scientist at Critical Technologies Inc., a doctoral candidate at Syracuse University, and an active member of too many professional and community organizations. The ultimate goal of his academic research is automatic generation, maintenance and exploitation of probably approximately correct, hybrid symbolic/numeric models of the world, self and other agents, for prediction, what-if analysis and control. The focus of his dissertation is development of an information theoretic framework for evolutionary learning.

Yossi Borenstein

He studied Computer Science at the University of Essex, U.K, where he received his Ph.D. in 2008. Currently he is a research fellow at the University of Hertfordshire. His PhD thesis, explores the connection between information theory and optimisation problems in the black box scenario. It consists of two parts: (1) a formal part which quantifies the amount of information that a landscape (problem in the black box scenario) contains and accordingly, using the notion of Kolmogorov complexity gives upper and lower bounds to performance, and (2) a part based on a linear approximation to the performance of a search algorithm which investigates the alignment of particular search algorithms with problems.

5

Graduate Student workshop

This full day workshop will involve presentations by approximately 10 selected students pursing research in some aspect of evolutionary computation. Students will make 15-20 minute presentations to an audience that will include a 'mentor' panel of established researchers in evolutionary computation. Presentations will be followed by a 10 minute question and discussion period led by the mentor panel.

Students will also be invited present their work as a poster at the evening Poster Session - an excellent opportunity to network with industry and academic members of the community.

The goal of this workshop is to assist students regarding their research: research methodology, goals, and plans. Students will also receive feedback on their presentation style. Other attendees will benefit by learning about current research, engaging in technical discussions and meeting researchers with related interests. Other students are encouraged to attend as a means of strengthening their own
research.

The group of presenting students will be chosen with the intent of creating a diverse group of students working on a broad range of topic areas.

Awards will be presented for best work and best presentation.

Workshop website

Riccardo Poli

He is a Professor in the School of Computer Science and Electronic Engineering at Essex. He started his academic career as an electronic engineer doing a PhD in biomedical image analysis to later become an expert in the field of EC. He has published around 250 refereed papers and two books on the theory and applications of genetic programming, evolutionary algorithms, particle swarm optimisation, biomedical engineering, brain-computer interfaces, neural networks, image/signal processing, biology and psychology. He is a Fellow of the International Society for Genetic and Evolutionary Computation (since 2003), a recipient of the EvoStar award for outstanding contributions to this field (2007), and an ACM SIGEVO executive board member (2007-2013). He was co-founder and co-chair of the European Conference on GP (1998-2000, 2003). He was general chair (2004), track chair (2002, 2007,2009), business committee member (2005), and competition chair (2006) of ACM's Genetic and Evolutionary Computation Conference, co-chair of the Foundations of Genetic Algorithms Workshop (2002) and technical chair of the International Workshop on Ant Colony Optimisation and Swarm Intelligence (2006). He is an associate editor of Genetic Programming and Evolvable Machines and the International Journal of Computational Intelligence Research. He is an advisory board member of Evolutionary Computation and the Journal on Artificial Evolution and Applications, and an editorial board member of Swarm Intelligence.

6

Thirteenth International Workshop on Learning Classifier Systems

Since Learning Classifier Systems (LCSs) were introduced by Holland [1] as a way of applying evolutionary computation to machine learning problems, the LCS paradigm has broadened greatly into a framework encompassing many representations, rule discovery mechanisms, and credit assignment schemes. Current LCS applications range from data mining to automated innovation to on-line control. Classifier systems are a very active area of research, with many approaches, in particular Wilson's accuracy-based XCS [2], receiving a great deal of attention. LCS are also benefiting from advances in reinforcement learning and other machine learning techniques.

This will be the 13th edition of the workshop, which was initiated in 1992, held at the NASA Johnson Space Center in Houston, Texas. Since 1999 the workshop has been held yearly in conjunction with PPSN in 2000 and 2002 and with GECCO in 1999, 2001 and from 2003 to 2009

Topics of interests include but are not limited to:

Paradigms of LCS (Michigan, Pittsburgh, ...)
Theoretical developments (behavior, scalability and learning bounds, ...)
Representations (binary, real-valued, oblique, non-linear, fuzzy, ...)
Types of target problems (single-step, multiple-step, regression/function approximation,...) - System enhancements (competent operators, problem structure identification and linkage learning, ...)
LCS for Cognitive Control (architectures, emergent behaviours, ...)
Applications (data mining, medical domains, bioinformatics, ...)

---------------------------------------
[1] J. H. Holland and J. S. Reitman.
Cognitive systems based on adaptive algorithms.
In D. Hayes-Roth and F. Waterman, editors, Pattern-directed Inference Systems, pages 313-329. Academic Press, New York, 1978.

[2] Steward W. Wilson.
Classifier fitness based on accuracy.
Evolutionary Computation, 3(2):149-175, 1995.
----------------------------------------

Workshop website

Jaume Bacardit

He received his Ph.D. in 2004 from the Ramon Llull University in Barcelona, Spain. His thesis studied the adaptation of the Pittsburgh approach of Learning Classifier Systems (LCS) to Data Mining tasks. In 2005 he joined the University of Nottingham, UK as a postdoctoral researcher to work on the application of LCS to data mine large-scale bioinformatics datasets and extract interpretable explanations from the learning process. In 2008 he was appointed as a Lecturer in Bioinformatics at the University of Nottingham. This is a joint post between the schools of Biosciences and Computer Science with the aim of developing interdisciplinary research at the interface of both disciplines. His research interests include the application of Learning Classifier Systems and other kinds of Evolutionary Learning to data mine large-scale challenging datasets and, in a general sense, the use of data mining and knowledge discovery for biological domains. He has been in the program committee, among other conferences and workshops, of the Genetic and Evolutionary Computation Conference (GECCO), the International Workshop on Learning Classifier Systems (IWLCS), the IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB) and the IEEE International Joint Conference on Neural Networks (IJCNN).

Bacardit has reviewed articles, among others, for the following journals: IEEE Transactions on System Man and Cybernetics Part B, IEEE Transactions on Evolutionary Computation, Evolutionary Computation, Evolutionary Intelligence, Natural Computation and Soft Computing. He has 28 refereed international publications between journal papers, conference papers and book chapters, and he has given 7 invited talks. Together with Dr. Martin Butz and Dr. Ester Bernadónsilla he co-organized the 2007 and 2008 editions of the IWLCS and co-edited a book with extended versions of the papers presented at IWLCS2006 and IWLCS2007.

Will Browne

He received the B. Eng. (Hons.) degree in Mechanical Engineering from the University of Bath, UK, in 1993 and the M. Sc. in Energy (Distinction) from the University of Wales, Cardiff in 1994. From 1994 to 1998 he was associated with British Steel and the University of Wales, Cardiff, through the Engineering Doctorate scheme, South Wales. His thesis regarded the industrial development of a Learning Classifier System for the Data Mining of quality control within a Steel Mill. From 1998 to 2001 he worked as a Post Doctoral Research Associate in the Control and Instrumentation Research Group, University of Leicester, UK. From October 2001 to August 2009 he lectured in the Cybernetic Intelligence Research Group (CIRG), University of Reading. He was appointed to Senior Lecturer, Victoria University of Wellington, NZ, in September 2009.

He has been involved with a wide range of EU projects. Currently, active in Fidis, Future of IDentity in the Information Society, FP6 NoE, a member of euCognition and EURON, EUropean RObotics research Network. His common research theme is developing systems capable of exploiting environmental feedback, which is core to both Cybernetics and Cognitive Robotics. He was an invited speaker at the Nokia Machine Consciousness workshop 2008 on artificial emotions for cognitive control of robotics. He presented an invited tutorial on Cognitive Robotics with Prof Kawamura (Vanderbilt University, USA) at Ro-man 2006. Conferences/Workshops Organisation has included PI of COGRIC: Cognitive Robotics and Control, EPSRC/NSF sponsored workshop 2006 that brought together internationally leading figures in order to discuss latest advancements and direct future research. Dr Browne also organised the 'Future Directions in Learning Classifier Systems' Workshop as part of PPSN 2004. He has been elected to serve on organising committee of International Workshop on Learning Classifier Systems for 2009 and 2010. Program Committees memberships /chairs include LCS and other GBML track at GECCO; the International Workshop on Learning Classifier Systems (IWLCS), Congress on Evolutionary Computation (CEC); Hybrid Intelligent Systems (HIS); Parallel Problem-Solving from Nature (PPSN) and Roman & Human Interactive Communication (RO-MAN). Journal reviewing has included IEEE Transactions on Evolutionary Computation, Journal of Soft Computing, Springer-Verlag, Journal of Engineering Manufacture, Journal of Pattern Analysis and Applications and IEEE Trans. Systems Man and Cybernetics. He has 29 refereed international publications between journal papers, conference papers and book chapters.

Jan Drugowitsch

He received his PhD in computer science at the University of Bath, UK, in October 2007 under the supervision of Alwyn Barry. His thesis "Learning Classifier Systems from First Principles: A Probabilistic Reformulation of Learning Classifier Systems from the Perspective of Machine Learning" reformulates Learning Classifier Systems (LCSs) from the point-of-view of statistical machine learning and provides an approach to tackle their design an analysis from the first principles of a statistical description of their aim. He is now working with Alexandre Pouget at the Department of Brain & Cognitive Sciences at the University of Rochester, USA, on the neural basis of decision making, with particular emphasis on multimodal integration and reaction time. He has been on the program committee for the Genetic and Evolutionary Computation Conference (GECCO) since 2005, the International Workshop on Learning Classifier Systems (IWLCS) since 2005, and has reviewed articles for the IEEE Transactions on Evolutionary Computation and the Memetic Computing journal. Drugowitsch has written the book "Design and Analysis of Learning Classifier Systems: A Probabilistic Approach", where he outlines the approach developed in his thesis and provides extensions to it.

7

Medical Applications of Genetic and Evolutionary Computation (MedGEC)

MedGEC 2010 is the sixth GECCO Workshop on the application of genetic and evolutionary computation (GEC) to problems in medicine and healthcare.

A dedicated workshop at GECCO provides a much needed focus for medical related applications of EC, not only providing a clear definition of the state of the art, but also support to practitioners for whom GEC might not be their main area of expertise or experience.

The Workshop has two main aims:
(i) to provide delegates with examples of the current state of the
art of applications of GEC to medicine.
(ii) to provide a forum in which researchers can discuss and
exchange ideas, support and advise each other in theory and practice.

Subjects will include (but are not limited to) applications of GEC to:

Medical imaging
Medical signal processing
Medical text analysis
Clinical diagnosis and therapy
Data mining medical data and records
Clinical expert systems
Modelling and simulation of medical processes
Drug description analysis
Patient-centric care

New for MedGEC 2010 will be a presentation detailing the range of medical applications to which GEC has been applied.

Workshop website

Stephen L. Smith

Received a BSc in Computer Science and then an MSc and PhD in Electronic Engineering from the University of Kent, UK. He is currently a senior lecturer in the Department of Electronics at the University of York, UK.

Steve's main research interests are in developing novel representations of evolutionary algorithms particularly with application to problems in medicine. His work is currently centred on the diagnosis of neurological dysfunction and analysis of mammograms. The former is currently undergoing clinical trials and being registered for a patent.

Steve was program chair for the Euromicro Workshop on Medical Informatics, program chair and local organizer for the Sixth International Workshop on Information Processing in Cells and Tissues (IPCAT) and guest editor for the subsequent special issue of BioSystems journal. Most recently he was tutorial chair for the IEEE Congress on Evolutionary Computation (CEC) in 2009. Steve is associate editor for Genetic Programming and Evolvable Machines journal, the Journal of Artificial Evolution and Applications, and a member of the editorial board for the International Journal of Computers in Healthcare.

Steve and Stefano are co-founders of the MedGEC Workshop, which is now in its sixth year. They are also guest editors for a special issue of Genetic Programming and Evolvable Machines on medical applications and preparing a book on the subject for publication in 2010.

Stefano Cagnoni
He is graduated in Electronic Engineering at the University of Florence in 1988 where he has been a PhD student and a post-doc until 1997. In 1994 he was a visiting scientist at the Whitaker College Biomedical Imaging and Computation Laboratory at the Massachusetts Institute of Technology. Since 1997 he has been with the University of Parma, where he has been Associate Professor since 2004. His main basic research interests concern soft computing, with particular regard to evolutionary computation, and computer vision. As concerns applied research, the main topics of his research are the application of the above-mentioned techniques to problems in computer vision, pattern recognition and robotics. Recent research grants regard: co-management of a project funded by Italian Railway Network Society (RFI) aimed at developing an automatic inspection system for train pantographs; a "Marie Curie" grant, for a four-year research training project in Medical Imaging using Bio-Inspired and Soft Computing; a grant from "Compagnia di S. Paolo" on "Bioinformatic and experimental dissection of the signalling pathways underlying dendritic spine function". He is Editor-in-chief of the "Journal of Artificial Evolution and Applications," has been chairman of EvoIASP since 1999, an event dedicated to evolutionary computation for image analysis and signal processing. Stefano is also a co-editor of special issues of journals dedicated to Evolutionary Computation for Image Analysis and Signal Processing and has been reviewer for international journals and member of the committees of several conferences. He is member of the Advisory Board of Perada, the UE Coordination Action on Pervasive Adaptation and has been recently awarded the "Evostar 2009 Award", in recognition of the most outstanding contribution to Evolutionary Computation.

Robert Patton

Dr. Patton received his Ph.D. in Computer Engineering with emphasis on Software Engineering from the University of Central Florida in 2002. In 2003, he joined the Applied Software Engineering Research group of Oak Ridge National Laboratory as a researcher. Dr. Patton primary research interests include data and event analytics, intelligent agents, computational intelligence, and nature-inspired computing. He currently is investigating novel approaches of evolutionary computation to the analysis of mammograms, abdominal aortic aneurysms, and traumatic brain injuries. In 2005, he served as a member of the organizing committee for the workshop on Ambient Intelligence - Agents for Ubiquitous Environments in conjunction with the 2005 Conference on Autonomous Agents and Multiagent Systems (AAMAS 2005).

8

Optimization by Building and Using Probabilistic Models (OBUPM-2010)

Genetic and evolutionary algorithms (GEAs) evolve a population of candidate solutions using two main operators: (1) selection and (2) variation. However, fixed, problem independent variation operators often fail to effectively exploit important features of high quality solutions obtained by selection to create novel, high-quality solutions. One way to make variation operators more effective is to replace traditional variation operators by the following two steps:

1. Estimate the distribution of the selected solutions on the basis of an adequate probabilistic model.

2. Generate a new population of candidate solutions by sampling from the distribution estimated.

Algorithms based on this principle are often called probabilistic model-building genetic algorithms (PMBGAs), estimation of distribution algorithms (EDAs) or iterated density estimation algorithms (IDEAs). The purpose of this workshop is to discuss recent advances in PMBGAs, theoretical and empirical results, applications of PMBGAs, and promising directions of future PMBGA research.

Workshop website

Mark Hauschild

He is a Ph.D. student in the department of Computer Science at the University of Missouri in St. Louis and a research assistant at the Missouri Estimation of Distribution Algorithms Laboratory (MEDAL). His research interests include efficiency enhancement of estimation of distribution algorithms (with a particular emphasis on exploiting prior knowledge), machine learning, scalability of evolutionary computation, probabilistic model building genetic programming and applying estimation of distribution algorithms to solve real-world problems.

Martin Pelikan

He received his Ph.D. from the Department of Computer Science at the University of Illinois at Urbana-Champaign in 2002. He joined the Department of Mathematics and Computer Science at the University of Missouri in St. Louis in 2003. Currently, he is an assistant professor of computer science and the director of the Missouri Estimation of Distribution Algorithms Laboratory (MEDAL). Pelikan's research focuses on genetic and evolutionary computation and machine learning. He worked at the Slovak University of Technology in Bratislava, the German National Center for Information Technology in Sankt Augustin, the Illinois Genetic Algorithms Laboratory (IlliGAL) at the University of Illinois at Urbana-Champaign, and the Swiss Federal Institute of Technology (ETH) in Zurich. Pelikan's most important contributions to genetic and evolutionary computation are the Bayesian optimization algorithm (BOA), the hierarchical BOA (hBOA), the scalability theory for BOA and hBOA, and efficiency enhancement techniques.

9

Symbolic Regression and Modeling Workshop

Symbolic Modeling is used to designate the search for symbolic descriptions, usually in the language of mathematics, to describe and predict numerical data in diverse fields such as industry, economics, finance and science.

Symbolic modeling captures the field of symbolic regression: a genetic programming based search technique for finding symbolic formulae on numerical data in order to obtain an accurate and concise description of that data in symbolic, mathematical form. In the evolutionary computation field it also captures learning classifier systems, if and when they are applied to obtain specific interpretable results in the field of interest.

Symbolic modeling can be defined as a set of techniques (including, but not limited to symbolic regression and learning classifier systems) and representations that try to find a mathematical description and prediction in some numerical space. This can be contrasted with numerical modeling such as (generalized) linear regression, neural networks, kernel regression and support vector machines.

The key discriminator of producing symbolic results over numerical results is the ability to interpret and analyze the results, leading either to acceptance by field experts, or to heightened understanding of the theory in the field of application. Interpretation is key, and the workshop will focus heavily on this.

The workshop will focus on advances in using symbolic modeling for real world problems in industry, economics, finance and science. Papers are sought that contribute to the state of the art in symbolic modeling, either through innovative applications, theoretical work on issues of generalization, size and comprehensibility of the results produced, and algorithmic improvements to make the techniques faster, more reliable and generally better controlled.

Workshop website

Steven Gustafson

He is a computer scientist at the General Electric Global Research Center in Niskayuna, New York. As a member of the Computational Intelligence Lab, he develops and applies advanced AI and machine learning algorithms for complex problem solving. He received his PhD in computer science from the University of Nottingham, UK, where he was a research fellow in the Automated Scheduling, Optimisation and Planning Research Group. He received his BS and MS in computer science from Kansas State University, where he was a research assistant in the Knowledge Discovery in Databases Laboratory. Dr. Gustafson is a member of several program committees, a member of the editorial board of the Journal of Artificial Evolution and Applications, and a Technical Editor-in-Chief of the new journal Memetic Computing. In 2006, he received the IEEE Intelligent System's "AI's 10 to Watch" award.

Mark Kotanchek:

Mark's diverse academic background (B.S. in Engineering Science, M.Eng. in Acoustics, Ph.D. in Aerospace Engineering, Senior Member of IEEE) is consistent with the diversity of professional experience (ladder design, defense system design, signal processing research, technology trend analysis, IT project management, chemical plant troubleshooting, high-throughput biology, energy trading system development, etc.). He is recognized as a global leader in genetic programming theory and application. He founded Evolved Analytics in 2005 with a goal of developing tools and systems to address the data deluge of the modern world and the need to convert that data into actionable insight and understanding.

10

Theoretical Aspects of Evolutionary Multiobjective Optimization - Current Status and Future Trends

Evolutionary Multiobjective Optimization (EMO), i.e., the simultaneous optimization of 2 or more objective functions by means of bio-inspired search heuristics, has become one of the main research fields in evolutionary computation in recent years and as such also gained interest from the classical field of multicriteria decision making (MCDM). Together with its fast development in practice, also theoretical analyses of EMO gained more and more interest. Besides several theoretical papers at international conferences such as GECCO or FOGA, also the current special issue of the Evolutionary Computation Journal devoted to "Theoretical Aspects of Evolutionary Multi-Objective Optimization" underpins this development.

To foster the growing interest of theory in EMO, the workshop aims at bringing together theoreticians both working in the field of EMO and from single-objective optimization as well as EMO practitioners. Besides presenting the newest theoretical results about EMO, the workshop aims at identifying a broad list of challenging questions to be tackled in the near future for which a collaboration between theoreticians and practitioners is expected to be highly beneficial. In this regard, we would like to invite both theoreticians and practitioners to talk and discuss about the latest developments in the field of theoretical analysis of evolutionary multiobjective optimization. It is highly appreciated if challenging questions and interesting new research directions are identified. The workshop covers all theoretical topics of EMO, including but not limited to

convergence analysis
decision making
diversity mechanisms
experimental studies based on theoretical considerations
hybridized EMO approaches
indicator-based search
many-objective optimization
multiobjectivization
performance assessment
preference handling
runtime analyses (incl. approximation results)
set-based EMO
variation operators specific to the multiobjective case

Workshop website

Dimo Brockhoff

He received his diploma in computer science from University of Dortmund, Germany in 2005 and his PhD (Dr. sc. ETH) from ETH Zurich, Switzerland in 2009. Since June 2009 he has been a postdoctoral researcher at INRIA Saclay Ile-de-France in Orsay, France in the TAO team of Michele Sebag and Marc Schoenauer. His research interests are focused on evolutionary multiobjective optimization (EMO), in particular on many-objective optimization as well as theoretical aspects of indicator-based search. Besides co-organizing the workshop on "Theoretical Aspects on Evolutionary Multiobjective Optimization", he also co-organizes the workshop "Theory of Randomized Search Heuristics" in March 2010 in Paris.

Nicola Beume

She received the Diploma degree in computer science from the University of Dortmund, Germany in 2006. Since then, she has been a Research Associate with the Faculty of Computer Science, Technische Universität Dortmund, Dortmund, Germany. Her research focuses on the design of multiobjective evolutionary algorithms as well as their theoretical and empirical analysis. She applies methods of computational intelligence to real-world problems, currently in the area of games.

11

Eighth GECCO Undergraduate Student Workshop

The eighth annual Undergraduate Student Workshop will occur on Wednesday, July 7, 2010. This half-day workshop will provide an excellent opportunity for undergraduate students to present a senior research project, summer project, or exceptional course project involving evolutionary computation, and receive valuable feedback from an international panel of experts. 6-page papers are due March 25; for additional details, please visit this website.

Clare Bates Congdon

She is an Assistant Professor of Computer Science and Chief Scientific Officer of Bioinformatics and Intelligent Systems at the University of Southern Maine. She received her BA from Wesleyan University and MS and PhD from the University of Michigan, and has been teaching evolutionary computation and machine learning to undergraduates for fifteen years. She is an advocate and mentor for undergraduate research, and has been bringing undergraduate students to GECCO and other conferences to present their evolutionary computation research since 2000. Her research (including that done with undergraduates) includes evolutionary computation as applied to areas such as bioinformatics, intelligent agents, and creative systems.

Frank Moore

He is an Associate Professor of Computer Science at the University of Alaska Anchorage. He received his BSCE, MSCE, and PhD from Wright State University. He has taught computer science, computer engineering, and electrical engineering courses at the undergraduate and graduate level since 1997. In addition, he has over six years of industry experience developing software for a wide variety of military projects. His recent research at the Air Force Research Laboratory has used evolutionary computation to optimize transforms that outperform wavelets for signal compression and reconstruction under conditions subject to quantization and thresholding.

12

1st GECCO Workshop on Visualization Methods for Genetic and Evolutionary Computation (VizGEC)

The first annual workshop on Visualization Methods for Genetic and Evolutionary Computation (VizGEC) is intended to explore and critically evaluate the development and application of visualization methods for enhancing the theoretical development of GEC and the application of GEC to real world problems. Visualization methods have an important role to play in understanding theoretical issues such as population structure, diversity and change over time as well as applied issues such as results interpretation and communication in a particular domain.

Jason H. Moore

Dr. Moore received his B.S. in Biological Sciences from Florida State University. He then received an M.S. in Human Genetics, an M.A. in Statistics and a Ph.D. in Human Genetics from the University of Michigan. He then served as Assistant Professor of Molecular Physiology and Biophysics (1999-2003) and Associate Professor of Molecular Physiology and Biophysics with tenure (2003-2004) at Vanderbilt University. While at Vanderbilt, Dr. Moore held an endowed position as an Ingram Associate Professor of Cancer Research. He also served as Director of the Bioinformatics Core and Co-Founder and Co-Director of the Vanderbilt Advanced Computing Center for Research and Education (ACCRE).

In 2004, Dr. Moore accepted a position as the Frank Lane Research Scholar in Computational Genetics, Associate Professor of Genetics, Associate Professor of Community and Family Medicine, and Director of Bioinformatics at Dartmouth Medical School. He was promoted to Full Professor with tenure in 2008. He also holds adjunct positions in the Department of Computer Science at the University of New Hampshire, the Department of Computer Science at the University of Vermont, the Department of Psychiatry at Brown University and the Translational Genomics Research Institute (TGen) in Phoenix. Dr. Moore serves as Director of the Computational Genetics Laboratory and the Bioinformatics Shared Resource for the Norris-Cotton Cancer Center, Director of the Integrative Biology Core for the Center for Environmental Health Sciences and Founder and Director of The DISCOVERY Resource, a 900-processor parallel computer cooperatively operated for the Dartmouth community. He also serves as Director of the newly established Bioinformatics Visualization Laboratory. His research has been communicated in more than 200 scientific publications and is supported by four NIH R01 grants in his name. He has previously served as Program Chair for the Bioinformatics and Computational Biology track at GECCO and as Chair of the European Conference on Evolutionary Computation, Machine Learning and Data Mining in Bioinformatics (EvoBIO).

13

Experimental Design and Statistical Analysis Workshop

This workshop is designed to be a hands-on companion to the Experimental Design and Statistics tutorials:

1. Statistical analysis for evolutionary computation (EC): Introduction (Mark Wineberg, Steffen Christensen)

2. Tuning and Experimental Analysis in EC: What We Still Have Wrong (Thomas Bartz-Beielstein, Mike Preuss)

This workshop will give you a chance to try out the various techniques learned in the tutorials in a controlled environment so you could then apply them to your own problems when you get back to your lab. Our field is becoming more and more rigorous in the application of appropriate experimental design and statistical analysis for any experiment being submitted the major conferences in the field. Make sure that you or your student’s have the experience you need to be able to perform a proper analysis. You supply your laptop; we will supply the tools and techniques. Parts of this workshop use “R”, a freely available language and environment for statistical computing and graphics. R provides a wide variety of statistical and graphical techniques. The software can be downloaded from http://cran.r-project.org.

The workshop will be presented at an introductory level and is meant for any EC researchers who want to compare their newly designed EC system with established EC systems to see if there is an improvement, or who want to determine which EC system performs better on their chosen problem; i.e., nearly everyone. It is vital, if our community is to be taken seriously, for us to continue to educate ourselves, and especially new Graduate students, on the statistical and experimental design techniques that are considered rigor for experiments performed in any field that is stochastic in nature, as is EC.

Topics Covered

1. Experimental Design: Factorial and space-filling designs, experimental setup and scope etc.

2. Statistical Analysis: Hypotheses testing, regression, analysis of variance (ANOVA), Bonferonni Post-hoc correction.

3. Sequential Parameter Optimization: Introduction to automatic and interactive tuning of algorithms. The sequential parameter optimization toolbox (SPOT) package for R is a toolbox for tuning and understanding simulation and optimization algorithms. SPOT includes methods for tuning based on classical regression and analysis of variance techniques; tree-based models such as CART and random forest; Gaussian process models (Kriging) and combinations of different meta-modeling approaches. This workshop exemplifies how SPOT can be used for automatic and interactive tuning.

Schedule

1. Intro Talk - orientation
2. Download relevant material. Installing an R system and the SPOT package on your computer
3. Design first experiment
4. Run first experiment
5. Analyze first experiment
6. Comparison: Interactive and automatic Tuning
7. Discussion, Q&A

References
SPOT can be downloaded from:
http://cran.r-project.org/web/packages/SPOT/index.html.
A SPOT related paper is available at http://arxiv.org/abs/1006.4645.
The new book “Experimental Methods for the Analysis of Optimization Algorithms” is also of interest, see http://www.springer.com/computer/ai/book/978-3-642-02537-2

Workshop website

Thomas Bartz-Beielstein

He is a professor for Applied Mathematics at Cologne University of Applied Sciences.

Steffen Christensen

He is a Senior Policy Researcher at the Policy Research Initiative.

Mike Preuss

Heis Research Associate at the Computer Science Department, University of Dortmund, Germany.

Mark Wineberg

He is Assistant Professor at the Department of Computing and Information Science, University of Guelph, Guelph, Ontario, Canada.