Benchmarking the (1+1) Evolution Strategy with One-Fifth Success Rule on the BBOB-2009 Function Testbed
Résumé
In this paper, we benchmark the (1+1) Evolution Strategy (ES) with one-fifth success rule which is one of the first and simplest adaptive search algorithms proposed for optimization. The benchmarking is conducted on the noise-free BBOB-2009 testbed. We implement a restart version of the algorithm and conduct for each run $10^{6}$ times the dimension of the search space function evaluations.
Domaines
Analyse numérique [cs.NA]Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...