An approach to credit assignment in classifier systems

@inproceedings{Westerdale1998AnAT,
  title={An approach to credit assignment in classifier systems},
  author={T. H. Westerdale},
  year={1998}
}
We present a conceptual framework within which we can analyze simple reward schemes for classifier systems. The framework consists of a set of classifiers, a learning mechanism, and a finite automaton environment that outputs payoff. We find that many reward schemes have negative biases that degrade system performance. We analyze bucket brigade schemes, which are subgoal reward schemes, and profit sharing schemes, which aren't. By contrasting these schemes, we hope to better understand the… CONTINUE READING

Topics from this paper.

Citations

Publications citing this paper.
SHOWING 1-10 OF 12 CITATIONS

Do Genes Have Fitness

VIEW 9 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Local Reinforcement and Recombination in Classifier Systems

  • Evolutionary Computation
  • 2001
VIEW 3 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Advances in Learning Classifier Systems

Pier Luca Lanzi, Wolfgang Stolzmann, Stewart W. Wilson
  • Lecture Notes in Computer Science
  • 2001
VIEW 1 EXCERPT
CITES BACKGROUND