This book provides a framework for the design of competent optimization techniques by combining advanced evolutionary algorithms with state-of-the-art machine learning techniques. The primary focus of the book is on two algorithms that replace traditional variation operators of evolutionary algorithms, by learning and sampling Bayesian networks: the Bayesian optimization algorithm (BOA) and the hierarchical BOA (hBOA). They provide a scalable solution to a broad class of problems. The book provides an overview of evolutionary algorithms that use probabilistic models to guide their search, motivates and describes BOA and hBOA in a way accessible to a wide audience, and presents numerous results confirming that they are revolutionary approaches to black-box optimization.

This book focuses like a laser beam on one of the hottest topics in evolutionary computation over the last decade or so: estimation of distribution algorithms (EDAs). EDAs are an important current technique that is leading to breakthroughs in genetic and evolutionary computation and in optimization more generally. I'm putting Scalable Optimization via Probabilistic Modeling in a prominent place in my library, and I urge you to do so as well. This volume summarizes the state of the art at the same time it points to where that art is going. Buy it, read it, and take its lessons to heart.

David E Goldberg, University of Illinois at Urbana-Champaign

This book is an excellent compilation of carefully selected topics in estimation of distribution algorithms---search algorithms that combine ideas from evolutionary algorithms and machine learning. The book covers a broad spectrum of important subjects ranging from design of robust and scalable optimization algorithms to efficiency enhancements and applications of these algorithms. The book should be of interest to theoreticians and practitioners alike, and is a must-have resource for those interested in stochastic optimization in general, and genetic and evolutionary algorithms in particular.

John R. Koza, Stanford University

This edited book portrays population-based optimization algorithms and applications, covering the entire gamut of optimization problems having single and multiple objectives, discrete and continuous variables, serial and parallel computations, and simple and complex function models. Anyone interested in population-based optimization methods, either knowingly or unknowingly, use some form of an estimation ofdistribution algorithm (EDA). This book is an eye-opener and a must-read text, covering easy-to-read yet erudite articles on established and emerging EDA methodologies from real experts in the field.

Kalyanmoy Deb, Indian Institute of Technology Kanpur

This book is an excellent comprehensive resource on estimation of distribution algorithms. It can serve as the primary EDA resource for practitioner or researcher. The book includes chapters from all major contributors to EDA state-of-the-art and covers the spectrum from EDA design to applications. These algorithms strategically combine the advantages of genetic and evolutionary computation with the advantages of statistical, model building machine learning techniques. EDAs are useful to solve classes of difficult real-world problems in a robust and scalable manner.

Una-May O'Reilly, Massachusetts Institute of Technology

Machine-learning methods continue to stir the public's imagination due to its futuristic implications. But, probability-based optimization methods can have great impact now on many scientific multiscale and engineering design problems, especially true with use of efficient and competent genetic algorithms (GA) which are the basis of the present volume. Even though efficient and competent GAs outperform standard techniques and prevent negative issues, such as solution stagnation, inherent in the older but more well-known GAs, they remain less known or embraced in the scientific and engineering communities. To that end, the editors have brought together a selection of experts that (1) introduce the current methodology and lexicography of the field with illustrative discussions and highly useful references, (2)exemplify these new techniques that dramatic improve performance in provable hard problems, and (3) provide real-world applications of these techniques, such as antenna design. As one who has strayed into the use of genetic algorithms and genetic programming for multiscale modeling in materials science, I can say it would have been personally more useful if this would have come out five years ago, but, for my students, it will be a boon.

Duane D. Johnson, University of Illinois at Urbana-Champaign

I'm not usually a fan of edited volumes. Too often they are an incoherent hodgepodge of remnants, renegades, or rejects foisted upon an unsuspecting reading public under a misleading or fraudulent title. The volume Scalable Optimization via Probabilistic Modeling: From Algorithms to Applications is a worthy addition to your library because it succeeds on exactly those dimensions where so many edited volumes fail. For example, take the title, Scalable Optimization via Probabilistic M- eling: From Algorithms to Applications. You need not worry that you're going to pick up this book and ?nd stray articles about anything else. This book focuseslikealaserbeamononeofthehottesttopicsinevolutionary compu- tion over the last decade or so: estimation of distribution algorithms (EDAs). EDAs borrow evolutionary computation's population orientation and sel- tionism and throw out the genetics to give us a hybrid of substantial power, elegance, and extensibility. The article sequencing in most edited volumes is hard to understand, but from the get go the editors of this volume have assembled a set of articles sequenced in a logical fashion. The book moves from design to e?ciency enhancement and then concludes with relevant applications. The emphasis on e?ciency enhancement is particularly important, because the data-mining perspectiveimplicitinEDAsopensuptheworldofoptimizationtonewme- ods of data-guided adaptation that can further speed solutions through the construction and utilization of e?ective surrogates, hybrids, and parallel and temporal decompositions.

This book provides a framework for the design of competent optimization techniques by combining advanced evolutionary algorithms with state-of-the-art machine learning techniques. The primary focus of the book is on two algorithms that replace traditional variation operators of evolutionary algorithms, by learning and sampling Bayesian networks: the Bayesian optimization algorithm (BOA) and the hierarchical BOA (hBOA). They provide a scalable solution to a broad class of problems. The book provides an overview of evolutionary algorithms that use probabilistic models to guide their search, motivates and describes BOA and hBOA in a way accessible to a wide audience, and presents numerous results confirming that they are revolutionary approaches to black-box optimization.