Do not match, inherit: Fitness surrogates for genetics-based machine learning techniques
28 March 2007Llorà X. Sastry K. Yu T.-L. Goldberg D. E. (2007). IlliGAL Report No. 2007011. University of Illinois at Urbana-Champaign, Urbana IL. [Full paper - PDF] [Full paper - PS].
Abstract:
One benefit of using probabilistic model-building genetic algorithms is the possibility of creating cheap and accurate surrogate models. Learning classifier systems—and genetics-based machine learning in general—can greatly benefit from such surrogates which can replace the costly matching procedure of a rule against large data sets. In this paper we investigate the accuracy of such surrogate fitness function when coupled with the probabilistic models evolved by the ?-ary extended compact classifier system (?eCCS). We present results showing how functional alignment between the probabilistic model of ?eCCS and the surrogate fitness is required. We also present a transformation of populations of rules based on the dependency structure matrix genetic algorithm (DSMGA) that allows building accurate models of overlapping building blocks—a necessary condition to accurately estimate the fitness of the evolved rules.
Powered by ScribeFire.
Related Posts:
- Do not match, Inherit: Fitness surrogates for genetics-based machine learning techniques
- Substructrual surrogates for learning decomposable classification problems: implementation and first results
- Automated alphabet reduction with evolutionary algorithms for protein structure prediction
- Don’t evaluate, inherit
- A survey of linkage learning techniques in genetic and evolutionary algorithms
No comments yet
