摘要: The optimization of a large number of decision variables, so called large scale global optimization (LSGO) remains challenging for existing heuristics. Inspired by the concept of global best (gbest) guided strategy, this paper pro... 展开 The optimization of a large number of decision variables, so called large scale global optimization (LSGO) remains challenging for existing heuristics. Inspired by the concept of global best (gbest) guided strategy, this paper proposes a gbest-guided covariance matrix adaptation evolution strategy (GCMA-ES) where the gbest information is utilized in the search equation to guide the exploitation process. The GCMA-ES can take advantages from both the CMA-ES and the gbest-guided strategy. Its performance is demonstrated on the CEC 2010 LSGO benchmarks. 收起