Prof.Dr. A.E. Eiben



Head of the Computational Intelligence Group
Department of Computer Science
Faculty of Sciences
Vrije Universiteit Amsterdam
T: +31-20-5987758
F: +31-20-5987653
E: gusz at cs.vu.nl

Address: Room T3.16, de Boelelaan 1081a, 1081HV Amsterdam, NL

Link to my old website (it has more links to papers in pdf)

SPOTLIGHT

My TED talk "Tech Kangaroos: Evolution at Work"
TEDx Danubia 2011, watch it here

The first comprehensive text book on evolutionary computing:
A.E. Eiben and J.E. Smith, Introduction to Evolutionary Computing, Springer,
First edition, 2003, ISBN 3-540-40184-9, Corrected 2nd printing, 2007, ISBN: 978-3-540-40184-1

The award winning paper on parameter control:
A.E. Eiben, R. Hinterding, and Z. Michalewicz, Parameter control in evolutionary algorithms, IEEE Transactions of EC, 3(2):124-141, 1999

The grand overview paper on parameter tuning:
A.E. Eiben and S.K. Smit, Parameter Tuning for Configuring and Analyzing Evolutionary Algorithms, Swarm and Evolutionary Computation, 1(1):19-31, 2011  

Vision paper about the (embodied) future of evolutionary computing:
A.E. Eiben, S. Kernbach, and Evert Haasdijk Embodied artificial evolution: Artificial evolutionary systems in the 21st Century, Evolutionary Intelligence, 2012, DOI: 10.1007/s12065-012-0071-x  

RESEARCH

My academic research lies within computational intelligence or natural computing with evolutionary computing as the binding factor, see the thematic overview on the right hand side of this page. I was/am involved in various European research projects: EvoNet I (Esprit 20996), EvoNet II (FP5, IST-1999-14087), DREAM (FP5, IST-1999-12679), NEW TIES (FP6-502386), SYMBRION (FP7-ICT-2007.8.2), EVOBODY (FP7-258334),OPTI-FOX (FP7-123456), AWARE (FP7-123456)

Further to academic research,  I have worked in business intelligence R&D projects, including data warehousing and data mining for feature selection, creditibility assessment, direct marketing, customer retention analysis, sensory data analysis, e-business, etc. Much of this work is unpublished.

TEACHING

Presently I teach the course Evolutionary Computing and Heuristics on the VU Amsterdam. I am also supervising Bachelor projects, Master Thesis projects and PhD students.

Previously, I have been teaching university courses on formal logic, search techniques, evolutionary computing, artificial life, evolutionary economy, and business intelligence on the Eindhoven University of Technology, the Utrecht University, Leiden University, and the VU Amsterdam. I was also lecturer of a business intelligence course for the e-Commerce Masters programme of the EUR, Rotterdam and for different business classes on the Universiteit Nyenrode, The Netherlands Business School.

ORGANIZATION AND MANAGEMENT

PUBLICATIONS

For a long list in chronological order click here. For a thematical overview see the blocks below on Theory, Multi-parent recombination, Constraint solving, Parameter tuning, Parameter control, Artificial life, Evolutionary robotics, Evolutionary Art, Evolution in time and space, and Miscellaneous (Under construction ...)

Multi-parent recombination

The "orgy in the computer" stuff. Using more than two parents can greatly improve (speed up) an evolutionary algorithm. Read more

Selected publications:
A.E. Eiben, Multiparent recombination, Handbook of EC

Parameter Control (on-line)

Changing EA parameter values during a run offers the greatest flexibility. Perhaps also the best performance. But it poses great challenges to EA designers. See the award winning paper that shaped the field. Read more

Selected publications:
A.E. Eiben, R. Hinterding, and Z. Michalewicz, Parameter control in evolutionary algorithms, IEEE Transactions of EC

Parameter Tuning (off-line)

There are many good EAs without parameter control, but there are no good EAs without well tuned parameters. Parameter tuning (before a run) is a must for good perfomance. See here the ins and outs of tuning and a good tuning toolbox Read more

Selected publications:
A.E. Eiben and S.K. Smit, Parameter Tuning for Configuring and Analyzing Evolutionary Algorithms, Journal of Swarm and Evolutionary Computing  

Artificial Life

Evolutionary algorithms can do more than solving optimization problems. They can propell emergent dynamics in a population of digital creatures. Evolution can also be combined with learning with sometimes surprising results. Read more

Selected publications:
T. Buresch T., A.E. Eiben, G. Nitschke, and M.C. Schut, Effects of Evolutionary and Lifetime Learning on Minds and Bodies in an Artifical Society, CEC 2005

Evolutionary Robotics

Designing robot controllers is a very complex task that can  greatly benefit from evolutionary approaches. Evolution of controllers is usually done off-line, but sure it can be done on-line too. Read more

Selected publications:
G.S. Nitschke, M.C. Schut, and A.E. Eiben, Collective neuro-evolution for evolving specialized sensor resolutions in a multi-rover task, Evolutionary Intelligence

Evolutionary Art

Interactive EAs where a human observer evaluates or directly selects objects created by mutation and crossover can deliver nice pieces of "art". And what about taking the human out of the loop? Read more

Selected publications:
A.E. Eiben, Evolutionary Reproduction of Dutch Masters: The Mondriaan and Escher Evolvers, The Art of Artificial Evolution: A Handbook on Evolutionary Art and Music

Evolution in Time and Space

Situated evolution, embodied evolution, distributed evolutionary algorithms, p2p evolution and the like. Can be much more challenging than a simple genetic algorithm solving the TSP...  Read more

Selected publications:
J.L.J. Laredo, A.E. Eiben, M. van Steen, and J. J. Merelo, EvAg: a scalable peer-to-peer evolutionary algorithm, Genetic Programming and Evolvable Machines

Theory

My early work in evolutionary computing. The first proof that  evolutionary algorithms (almost) always converge to a solution. Read more

Selected publications:
A.E. Eiben, E.H.L. Aarts, and K.M. van Hee, Global convergence of genetic algorithms: A Markov chain analysis, PPSN 1990

Constraint solving

Evolutionary algorithms are usually blind to constraints. But many challenging problems are constrained. And there are  some good tricks... Read more

Selected publications:
A.E. Eiben, J.K. van der Hauw, and J.I. van Hemert. Graph coloring with adaptive evolutionary algorithms, Journal of Heuristics

Miscellaneous

Various papers on research methodology, general intoduction to evolutionary computing, mathematical logic, etc  ...  Read more

Selected publications:
A.E. Eiben and M. Jelasity, A Critical Note on Experimental Research Methodology in EC, CEC 2002