Adaptation in natural and artificial systems
Adaptation in natural and artificial systems
Evolutionary algorithms in theory and practice: evolution strategies, evolutionary programming, genetic algorithms
Genetic algorithms + data structures = evolution programs (3rd ed.)
Genetic algorithms + data structures = evolution programs (3rd ed.)
Bimodal performance profile of evolutionary search and the effects of crossover
Theoretical aspects of evolutionary computing
Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Uniform Crossover in Genetic Algorithms
Proceedings of the 3rd International Conference on Genetic Algorithms
Modeling Building-Block Interdependency
PPSN V Proceedings of the 5th International Conference on Parallel Problem Solving from Nature
PPSN V Proceedings of the 5th International Conference on Parallel Problem Solving from Nature
The science of breeding and its application to the breeder genetic algorithm (bga)
Evolutionary Computation
Hi-index | 0.00 |
Previous publications by the authors have demonstrated a bimodal performance profile for simple evolutionary search on variants of the Adaptive Distributed Database Management Problem (ADDMP) and other problems over a range of evaluation limits. This paper examines an anomaly seen in one of these profiles and together with results from a range of other problems, shows that with sufficiently high evaluation limits, a multimodal performance profile is apparent in search spaces with significant numbers of deceptive local optima. This is particularly apparent in the performance profile of the Hierarchial If and only If problem (H-IFF) where the regular structure of the search space produces several distinct peaks and troughs in the performance profile, possibly indicative of a range of specific 'fitness barriers' which are surmountable by specific rates of mutation. This observation could prove important in general EA parameter tuning over a range of problems with similar characteristics. Further, the existence of optimal mutation rates inducing a minimum in standard deviation of run-time, is of critical importance in the application of EAs to real-time, real-world problems.