Step Size Adaptation in Evolution Strategies using Reinforcement Learning

S. Müller, N. N. Schraudolph, and P. D. Koumoutsakos. Step Size Adaptation in Evolution Strategies using Reinforcement Learning. In Proc. Congress on Evolutionary Computation, pp. 151–156, IEEE, 2002.

Download

pdf djvu ps.gz
75.4kB   73.9kB   203.9kB  

Abstract

We discuss the implementation of a learning algorithm for determining adaptation parameters in evolution strategies. As an initial test case, we consider the application of reinforcement learning for determining the relationship between success rates and the adaptation of step sizes in the (1+1)-evolution strategy. The results from the new adaptive scheme when applied to several test functions are compared with those obtained from the (1+1)-evolution strategy with a priori selected parameters. Our results indicate that assigning good reward measures seems to be crucial to the performance of the combined strategy.

BibTeX Entry

@inproceedings{MueSchKou02,
     author = {Sybille M\"uller and Nicol N. Schraudolph
               and Petros D. Koumoutsakos},
      title = {\href{http://nic.schraudolph.org/pubs/MueSchKou02.pdf}{
               Step Size Adaptation in Evolution Strategies
               using Reinforcement Learning}},
      pages = {151--156},
     volume =  1,
  booktitle = {Proc.\ Congress on Evolutionary Computation},
  publisher = {IEEE},
       year =  2002,
   b2h_type = {Other},
  b2h_topic = {Evolutionary Algorithms},
   abstract = {
    We discuss the implementation of a learning algorithm for determining
    adaptation parameters in evolution strategies. As an initial test case,
    we consider the application of reinforcement learning for determining
    the relationship between success rates and the adaptation of step sizes
    in the (1+1)-evolution strategy. The results from the new adaptive scheme
    when applied to several test functions are compared with those obtained
    from the (1+1)-evolution strategy with a priori selected parameters.
    Our results indicate that assigning good reward measures seems to be
    crucial to the performance of the combined strategy.
}}

Generated by bib2html.pl (written by Patrick Riley) on Thu Sep 25, 2014 12:00:33