An Evolutionary Many-Objective Optimization Algorithm Based on Coverage and Cache Strategy
Haoran Sun, Xinye Cai, Muhammad Sulaman, Zhun Fan · 2017
How to balance the diversity and convergence plays an important role on the performance of a multiobjective evolutionary optimizer. Due to the loss of selection pressure and the exponential expansion in the high-dimensional objective space, it is even more difficult for an optimizer to balance between convergence and diversity for a many-objective optimization problem. To address this issue, in this paper, we propose a cache mechanism to improve the convergence and a coverage-based method for maintaining better diversity. Based on these two mechanisms, a many-objective evolutionary algorithm is further proposed. The experimental studies are conducted to verify the effectiveness of the proposed approach.