Advertisement

Table of contents

  1. Front Matter
  2. Martin Pelikan
    Pages 31-48
  3. Martin Pelikan
    Pages 49-87
  4. Martin Pelikan
    Pages 89-103
  5. Martin Pelikan
    Pages 105-129
  6. Martin Pelikan
    Pages 131-146
  7. Martin Pelikan
    Pages 147-149
  8. Back Matter

About this book

Introduction

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 book focuses 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). BOA and hBOA are theoretically and empirically shown to provide robust and scalable solution for broad classes of nearly decomposable and hierarchical problems. A theoretical model is developed that estimates the scalability and adequate parameter settings for BOA and hBOA. The performance of BOA and hBOA is analyzed on a number of artificial problems of bounded difficulty designed to test BOA and hBOA on the boundary of their design envelope. The algorithms are also extensively tested on two interesting classes of real-world problems: MAXSAT and Ising spin glasses with periodic boundary conditions in two and three dimensions. Experimental results validate the theoretical model and confirm that BOA and hBOA provide robust and scalable solution for nearly decomposable and hierarchical problems with only little problem-specific information.

Keywords

Analysis Bayesian network algorithm algorithms evolutionary algorithm genetic algorithms learning machine learning modeling operator optimization

Bibliographic information

  • DOI https://doi.org/10.1007/b10910
  • Copyright Information Springer-Verlag Berlin/Heidelberg 2005
  • Publisher Name Springer, Berlin, Heidelberg
  • eBook Packages Engineering
  • Print ISBN 978-3-540-23774-7
  • Online ISBN 978-3-540-32373-0
  • Series Print ISSN 1434-9922
  • Series Online ISSN 1860-0808
  • Buy this book on publisher's site