Parallel genetic algorithms. Theory and real world applications

DOI10.1007/978-3-642-22084-5MaRDI QIDQ2275411zbMATH OpenOpenAlexFDO

Publication date 8 August 2011

Published in Studies in Computational Intelligence (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-3-642-22084-5

evolutionary algorithms parallel evolutionary algorithms

Research exposition (monographs, survey articles) pertaining to computer science (68-02) Problem solving in the context of artificial intelligence (heuristics, search strategies, etc.) (68T20) Parallel algorithms in computer science (68W10) Distributed algorithms (68W15)

Review text

Evolutionary algorithms are a large and diverse class of randomized search heuristics that are successfully applied in practice for problems where no good problem-specific algorithm is available. Their main advantages are that they are easy to implement and apply to a new problem and often deliver good solutions with acceptable computational effort. One popular way of further increasing their practicability is decreasing the wall-clock time needed to find good solutions by means of parallelization. This important practical aspect is the book's topic. The book consists of three parts and an appendix. The first part briefly introduces the main notions and notations, different types of parallel evolutionary algorithms, and a guide for performing rigorous experimental research in the area including the presentation of the results in a scientific context. While the introduction of the terminology and notation appears a bit rushed and at times erroneous, the presentation of the different algorithms demonstrates an excellent overview of the field. Moreover, the guide towards experimentation and presentation of result is an important contribution to the field. The second part is devoted to the more theoretical aspects and contains only one chapter. It sketches existing models for parallel evolutionary algorithms and concentrates on migration and its influence on growth curves for the proportion of best individuals in the population and, consequently, on the take over time. The third and longest part contains four case studies and empirical comparisons of parallel evolutionary algorithms with other randomized search heuristics, touching the fields of natural language processing, design of digital logic circuits, workforce planning, and the assembly of DNA fragments from the field of bioinformatics. Finally, the appendix describes a library that helps with the implementation of parallel evolutionary algorithms. While the subtitle `Theory and real world applications' suggests a broad and balanced view of the field, the book is clearly addressed at the practitioner in the field and at people interested in applying parallel evolutionary algorithms to some specific real world problem. People more interested in theory will find its scope limited and will further be irritated by a relatively large number of small (mostly notational) errors. For practitioners, the book offers four very concrete and precise case studies, an introduction to a library that facilitates implementation and a thorough and practical guide towards presenting their own results in scientific publications. The book extends well beyond a summary of the state of the art and contains a large amount of original research results.

Recommendations

Cited in

(28)

This page was built for publication: Parallel genetic algorithms. Theory and real world applications

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2275411)