Multi-sample Evolution of Robust Black-Box Search AlgorithmsProceedings of the 2013 Genetic and Evolutionary Computation Conference Companion (GECCO 2014)
AbstractBlack-Box Search Algorithms (BBSAs) tailored to a specific problem class may be expected to significantly outperform more general purpose problem solvers, including canonical evolutionary algorithms. Recent work has introduced a novel approach to evolving tailored BBSAs through a genetic programming hyper-heuristic. However, that first generation of hyper-heuristics suffered from overspecialization. This poster paper presents a second generation hyperheuristic employing a multi-sample training approach to alleviate the overspecialization problem. A variety of experiments demonstrated the significant increase in the robustness of the generated algorithms due to the multi-sample approach, clearly showing its ability to outperform established BBSAs. The trade-off between a priori computational time and the generated algorithm robustness is investigated, demonstrating the performance gain possible given additional run-time.
Meeting Name16th Genetic and Evolutionary Computation Conference, GECCO 2014 (2014: Jul. 12-16, Vancouver, British Columbia, Canada)
Research Center/Lab(s)Center for High Performance Computing Research
Sponsor(s)Missouri University of Science and Technology. Natural Computation Laboratory
Keywords and Phrases
- Black-Box Search Algorithms,
- Evolutionary Algorithms,
- Genetic Programming,
International Standard Book Number (ISBN)9781450328814
Document TypeArticle - Conference proceedings
Rights© 2014 Association for Computing Machinery (ACM), All rights reserved.
Citation InformationMatthew A. Martin and Daniel R. Tauritz. "Multi-sample Evolution of Robust Black-Box Search Algorithms" Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion (GECCO 2014) (2014) p. 195 - 196
Available at: http://works.bepress.com/daniel-tauritz/52/