Black-Box Optimization Benchmarking of Two Variants of the POEMS Algorithm on the Noiseless Testbed

摘要

This paper presents benchmarking of a stochastic local search algorithm called Prototype Optimization with Evolved Improvement Steps (POEMS) on the BBOB 2010 noise-free functions testbed. An original version of the POEMS algorithm presented at BBOB 2009 workshop is compared to a new variant using a pool of candidate prototypes. Experiments for 2D, 3D, 5D, 10D and 20D were done. Experimental results show that the new variant of POEMS performs better on several functions for lower dimensions. Both variants perform equally on the 20D problems.