Adaptive Large-Scale Multi-Objective Evolutionary Optimization Based on Reference Solution Guidance

Abstract

The decision space of large-scale multi-objective evolutionary optimization problems is broader, which makes the solving process more difficult. In this paper, we propose an adaptive large-scale multi-objective optimization algorithm based on reference solution guidance. The algorithm uses a cyclic selection strategy to screen the population and an adaptive generation strategy to generate offspring solutions. Finally, a decomposition-based dual environmental selection strategy is used to improve the quality of the population. We compared the proposed algorithm with other common large-scale multi-objective optimization algorithms. The experimental results show that this algorithm has excellent performance and effectiveness and can effectively solve large-scale multi-objective optimization problems.