Abstract | ||
---|---|---|
The exploration vs exploitation dilemma is to balance exploring new but potentially less.t regions of the.tness landscape while also focusing on regions near the.ttest individuals. For the tunable problem class SPARSELOCALOPT a non-elitist EA with tournament selection can limit the percentage of "sparse" local optimal individuals in the population using a su.ciently high mutation rate (Dang et al., 2021). However, the performance of the EA depends critically on choosing the "right" mutation rate, which is problem instance-speci.c. A promising approach is self-adaptation, where parameter settings are encoded in chromosomes and evolved. We propose a new self-adaptive EA for single-objective optimisation, which treats parameter control from the perspective of multiobjective optimisation: The algorithm simultaneously maximises the.tness and the mutation rates. Since individuals in "dense".tness valleys survive high mutation rates, and individuals on "sparse" local optima only survive with lower mutation rates, they can coexist on a non-dominated Pareto front. Runtime analyses show that this new algorithm (MOSA-EA) can e.ciently escape a local optimum with unknown sparsity, where some.xed mutation rate EAs become trapped. Complementary experimental results show that the MOSA-EA outperforms a range of EAs on random NK.L-ANDSCAPE and k-S-AT instances. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1145/3512290.3528836 | PROCEEDINGS OF THE 2022 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO'22) |
Keywords | DocType | Citations |
Evolutionary algorithms, self-adaptation, multi-modal functions | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Per Kristian Lehre | 1 | 627 | 42.60 |
Xiaoyu Qin | 2 | 0 | 0.68 |