论文信息 - Global Selection Methods for Massively Parallel Computers

Global Selection Methods for Massively Parallel Computers

In this paper, the parallel implementation of various selection operators in evolutionary algorithms on massively parallel 2-dimensional arrays is treated. Novel parallel versions of fitness proportionate, linear ranking, and tournament selection are presented and compared. It is found that these algorithms can be implemented such that their expected behaviour is identical or very close to those of the sequential algorithms from which they originated, but with the advantage of significant speed improvements associated with parallelism.

Jürgen Branke | Hartmut Schmeck | Hans Christian Andersen

[1] Kalyanmoy Deb,et al. A Comparative Analysis of Selection Schemes Used in Genetic Algorithms , 1990, FOGA.

[2] Thomas Bäck,et al. Selective Pressure in Evolutionary Algorithms: A Characterization of Selection Mechanisms , 1994, International Conference on Evolutionary Computation.

[3] Denis Trystram,et al. Parallel algorithms and architectures , 1995 .

[4] James E. Baker,et al. Reducing Bias and Inefficienry in the Selection Algorithm , 1987, ICGA.

[5] David R. Jefferson,et al. Selection in Massively Parallel Genetic Algorithms , 1991, ICGA.

[6] Kenneth A. De Jong,et al. On Decentralizing Selection Algorithms , 1995, ICGA.

[7] L. Darrell Whitley,et al. GENITOR II: a distributed genetic algorithm , 1990, J. Exp. Theor. Artif. Intell..

[8] Peter J. B. Hancock,et al. An Empirical Comparison of Selection Methods in Evolutionary Algorithms , 1994, Evolutionary Computing, AISB Workshop.

[9] Heinz Mühlenbein,et al. Evolution in Time and Space - The Parallel Genetic Algorithm , 1990, FOGA.

[10] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .