I want to inform the UAI community of a new book Estimation of Distribution Algorithms. A new Tool for Evolutionary Computation.
Edited by P. Larra�aga and Jose A. Lozano, Kluwer Academic Publishers, 2002. For more detailed information, table of contents, abstracts and chapters see: http://www.sc.ehu.es/isg http://www.wkap.nl/prod/s/GENA Summary: The book is devoted to a new paradigm for Evolutionary Computation, named Estimation of Distribution Algorithms (EDAs). Based on Genetic Algorithms (GAs), this new class of algorithms generalizes GAs by replacing the crossover and mutation operators by learning and sampling the probability distribution of the best individuals of the population at each iteration of the algorithm. Working in such a way, the relationships between the variables involved in the problem domain are explicitly and effectively captured and exploited. This text constitutes the first compilation and review of the techniques and applications of this new tool for performing Evolutionary Computation. The book is clearly divided into three parts and comprised of a total of 18 chapters. Part I is dedicated to the foundations of EDAs. In this part different paradigms for Evolutionary Computation are introduced and some probabilistic graphical models --Bayesian networks and Gaussian networks-- used in learning and sampling the probability distribution of the selected individuals at each iteration of EDAs are presented. In addition to this, a review of the existing EDA approaches is carried out. Also EDAs based on the learning mixture models are presented and some approaches to the parallelization of the learning task are introduced. This part concludes with the mathematical modeling of some of the proposed EDA approaches. Part II brings together several applications of EDAs in optimization problems and reports on the results reached. Among the solved problems are the following ones: the traveling salesman problem, the job scheduling problem and the knapsack problem, as well as the optimization of some well-known combinatorial and continuous functions. This part ends with a chapter devoted to an EDA based approach to the inexact graph matching problem. Part III presents the application of EDAs in order to solve some problems that arise in the Machine Learning field. Concretely, the problems considered are: feature subset selection, feature weigthing in K-NN classifiers, rule induction, partial abductive inference in Bayesian networks, partitional clustering and the searching for optimal weights in artificial neural networks. This book can be a useful and interesting tool for researchers working in the field of Evolutionary Computation. Also engineers who, in their every day life, face real-world optimization problems and whom are provided with a new and powerful tool can derive benefit from the reading of the book. Moreover, this book may be used by graduate students in computer science and by people interested in taking part in the development of this new methodology that, in the following years, will provide us with interesting and appealing challenges. (.~.) -----------------------------------------oOO--(_)--OOo------------- J.A. Lozano e-mail: [EMAIL PROTECTED] Computer Science & AI Department tfno.: +34-943-015034 University of the Basque Country http://www.sc.ehu.es/isg Aptdo. 649 20080 San Sebastian (Spain) -------------------------------------------------------------------
