For many practical optimization problems, the evaluation of a solution is subject to noise, and optimization heuristics capable of handling such noise are needed. In this paper, we examine the influence of noise on particle swarm optimization and demonstrate that the resulting stagnation can not be removed by parameter optimization alone, but requires a reduction of noise through averaging over multiple samples. In order to reduce the number of required samples, we propose a combination of particle swarm optimization and a statistical sequential selection procedure, called optimal computing budget allocation, which attempts to distribute a given number of samples in the most effective way. Experimental results show that this new algorithm indeed outperforms the other alternatives.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.