Benchmarking the (1+1) Evolution Strategy with One-Fifth Success Rule on the BBOB-2009 Function Testbed

Anne Auger (1, 2)

1 TAO - Machine Learning and Optimisation
2 MSR - INRIA - Microsoft Research - Inria Joint Centre

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]

Anne Auger. Benchmarking the (1+1) Evolution Strategy with One-Fifth Success Rule on the BBOB-2009 Function Testbed. ACM-GECCO Genetic and Evolutionary Computation Conference, Jul 2009, Montreal, Canada. ⟨inria-00430515⟩