论文信息 - Evolution strategies in noisy environments- a survey of existing work

Evolution strategies in noisy environments- a survey of existing work

Noise is a factor present in almost all real-world optimization problems. While it can potentially improve convergence reliability in multimodal optimization by preventing convergence towards merely local optima, it is generally detrimental to the velocity with which an optimum is approached. Evolution strategies (ES) form a class of evolutionary optimization procedures that are believed to be able to cope quite well with noise. A number of theoretical results as well as empirical findings regarding the influence of noise on the performance of ES can be found in the literature. The purpose of this survey is to summarize what is known regarding the behavior of ES in noisy environments and to outline directions for future research.

Dirk V. Arnold | D. Arnold

[1] Hans-Georg Beyer,et al. Toward a Theory of Evolution Strategies: On the Benefits of Sex the (/, ) Theory , 1995, Evolutionary Computation.

[2] Hans-Georg Beyer,et al. Toward a Theory of Evolution Strategies: The (, )-Theory , 1994, Evolutionary Computation.

[3] Günter Rudolph,et al. Local convergence rates of simple evolutionary algorithms with Cauchy mutations , 1997, IEEE Trans. Evol. Comput..

[4] Volker Nissen,et al. Optimization with Noisy Function Evaluations , 1998, PPSN.

[5] Hans-Paul Schwefel,et al. Where Elitists Start Limping Evolution Strategies at Ridge Functions , 1998, PPSN.

[6] H. Beyer. Evolutionary algorithms in noisy environments : theoretical issues and guidelines for practice , 2000 .

[7] Hillol Kargupta,et al. A perspective on the foundation and evolution of the linkage learning genetic algorithms , 2000 .

[8] Hans-Georg Beyer,et al. Toward a Theory of Evolution Strategies: Some Asymptotical Results from the (1,+ )-Theory , 1993, Evolutionary Computation.

[9] H.-G. Beyer,et al. Mutate large, but inherit small ! On the analysis of rescaled mutations in (1, λ)-ES with noisy fitness data , 1998 .