Titel: Analyzing Evolutionary Algorithms
Autor/en: Thomas Jansen
The Computer Science Perspective.
HC runder Rücken kaschiert.
Springer Berlin Heidelberg
19. Dezember 2012 - gebunden - 268 Seiten
Evolutionary algorithms is a class of randomized heuristics inspired by natural evolution. They are applied in many different contexts, in particular in optimization, and analysis of such algorithms has seen tremendous advances in recent years.
In this book the author provides an introduction to the methods used to analyze evolutionary algorithms and other randomized search heuristics. He starts with an algorithmic and modular perspective and gives guidelines for the design of evolutionary algorithms. He then places the approach in the broader research context with a chapter on theoretical perspectives. By adopting a complexity-theoretical perspective, he derives general limitations for black-box optimization, yielding lower bounds on the performance of evolutionary algorithms, and then develops general methods for deriving upper and lower bounds step by step. This main part is followed by a chapter covering practical applications of these methods.
The notational and mathematical basics are covered in an appendix, the results presented are derived in detail, and each chapter ends with detailed comments and pointers to further reading. So the book is a useful reference for both graduate students and researchers engaged with the theoretical analysis of such algorithms.
Introduction; Evolutionary Algorithms and Other Randomized Search Heuristics; Theoretical Perspectives on Evolutionay Algorithms; General Limits in Black-Box Optimization; Methods for the Analysis of Evolutionary Algorithms; Selected Topics in the Analysis of Evolutionary Algorithms; App. A, Landau Notation; App. B, Tail Estimations; App. C, Martingales and Applications
The author lectured and researched in the Technische Universität Dortmund for 9 years after his PhD, and he is now the Stokes College Lecturer in the Department of Computer Science in University College Cork. He has tested the book content in his own lectures at these universities, and he has been invited to run the tutorial on this subject at the main international conference on evolutionary computing, GECCO.
"[The book] is aimed at evolutionary computation researchers and enthusiasts who are interested in the theoretical analysis of evolutionary algorithms. [It] will be accessible to post-graduates and advanced undergraduates in mathematics and/or computer science, and generally anyone with a working background in discrete mathematics, algorithms, and basic probability theory. Theoreticians will benefit from this book because it works well as a convenient reference for essential analytical strategies and many up-to-date results. Students and newcomers to the field will find the book a handy compendium of techniques that have become indispensable for writing runtime proofs in the evolutionary computation theory community such as black box complexity, drift analysis, fitness-based partitions, and the method of typical runs. Each of these techniques is carefully motivated and presented in a manner that is suitably rigorous, yet again not needlessly arcane. Additionally, at the end of each chapter, there are useful citations to contemporary work that provide more detailed treatments of many of the topics presented in the chapter. Practitioners can also benefit from [the book] since the theoretical foundations it presents serve to illuminate the working principles behind evolutionary algorithms and offer insights into the random processes that govern their behavior. Thus it can cultivate a deeper understanding of these aspects, and such an understanding is crucial for an informed approach to the design of good algorithms."[A.M. Sutton, Genetic Programming and Evolvable Machines (2013) 14:473-475]