Deep reinforcement learning-based two-stage coevolutionary framework for multimodal multi-objective optimization

doi:10.1016/j.asoc.2026.114836

Deep reinforcement learning-based two-stage coevolutionary framework for multimodal multi-objective optimization

2026-02-18

0

PRE

AI

DOI：10.1016/j.asoc.2026.114836

Original

Original request for help

Save

Abstract

En 中文

• Deep reinforcement learning based adaptive evolutionary operator selection strategy. • Incorporating adaptive operators into a two-stage co-evolutionary framework proposing MMOEA-DRL. • MMOEA-DRL achieves 64.71% optimality and outperforms other methods by more than 55.74% in the decision space.

Keywords:

Deep reinforcement learning

Adaptive evolutionary operators

Two-stage co-evolutionary framework

Multimodal multi-objective optimization

Evolutionary algorithms

Deep reinforcement learning-based two-stage coevolutionary framework for multimodal multi-objective optimization

Abstract

Journal

Organization