Session: Heuristics, Metaheuristics and Hyper-Heuristics II (06/08, 11:15-13:15, Room 7)

Clustering-Based Evolution Control for Surrogate-Assisted Particle Swarm Optimization

Clustering-Based Evolution Control for Surrogate-Assisted Particle Swarm Optimization

Haibo Yu, Ying Tan, Chaoli Sun and Jianchao Zeng

Haibo Yu, Ying Tan, Chaoli Sun and Jianchao Zeng

Taiyuan University of Science and Technology, China Taiyuan University of Science and Technology, China Taiyuan University of Science and Technology, China North University of China, China

Taiyuan University of Science and Technology, China Taiyuan University of Science and Technology, China Taiyuan University of Science and Technology, China North University of China, China

When using a fixed number of neighbors for training a local surrogate model in surrogate assisted evolutionary optimization algorithms, it may suffer from the large uncertainty because the actual distribution of candidate's neighborhood may be neglected. In this paper, we propose to firstly analyze the distribution characteristics of candidate's neighborhood through a modified overlapping clustering method before training a local surrogate, and then use the clustering based evolution control strategy or model management strategy to facilitate the evolutionary algorithm to converge to the right optimum. Simulation results on four widely used benchmark functions demonstrate the efficacy of the proposed method.