"A Computational Intelligence Framework for
Problem Solving through Co-evolution"
by Siang Yew CHONG

Abstract

For a computational system in problem solving to exhibit intelligence, it must be able to adapt its behavior to predict future outcomes and take the appropriate decision over a range of environments. Here, co-evolutionary computation has been introduced as an attractive framework to address these issues of intelligence through its emphasis of adaptability in decision-making. This talk will introduce co-evolutionary computation and present broad and fundamental issues of the framework in the context of strategic decision-making. In particular, the presentation will discuss how the framework can be formulated as a general search methodology, e.g., for optimization and learning [1], and also as tools for simulation and modeling purposes [2]. A fundamental issue of co-evolutionary computation studies and applications is that of robustness (performance) of solutions obtained. The talk will introduce a recent theoretical advancement we made in [3] to address the issue of rigorous quantitative performance analysis in co-evolutionary computation and discuss its various practical implications through computational studies. Examples in games such as the prisoner's dilemma will be used throughout the talk to illustrate and highlight these issues.

[1] S. Y. Chong, M. K. Tan, and J. D. White, "Observing the Evolution of Neural Networks Learning to Play the Game of Othello,'' IEEE Transactions on Evolutionary Computation, Vol. 9, No. 3, pp. 240-251, Jun. 2005.

[2] S. Y. Chong and X. Yao, "Multiple Choices and Reputation in Multiagent Interactions,'' IEEE Transactions on Evolutionary Computation, Vol. 11, No. 6, pp. 689-711, Dec. 2007.

[3] S. Y. Chong, P. Tino, and X. Yao, "Measuring Generalization Performance in Coevolutionary Learning,'' IEEE Transactions on Evolutionary Computation, In Press.

About the speaker

Siang Yew CHONG received his Ph.D. in Computer Science from the University of Birmingham, UK, in 2007. He was formerly a Research Associate with the Centre of Excellence for Research in Computational Intelligence and Applications (Cercia), a member of the Natural Computation Research Group, and is currently an Honorary Research Fellow with the School of Computer Science, University of Birmingham, UK.

His main research interests include broad areas in computational intelligence such as evolutionary computation and neural networks, machine learning, evolutionary game theory, and their various applications in problem solving, simulation and modeling (e.g., in strategic decision-making using games such as the prisoner's dilemma). He regularly contributes to studies in computational intelligence such as the IEEE Transactions on Evolutionary Computation, which is published by the IEEE Computational Intelligence Society. He has recently co-edited the book, The Iterated Prisoners' Dilemma: 20 Years On, with Graham Kendall and Xin Yao, which highlights the latest research since the seminal work by Robert Axelrod. He also reviews for a number of leading journals, books and conferences in computational intelligence.