A deep reinforcement learning-guided multimodal multi-objective evolutionary algorithm with a serial-parallel mechanism

Ying Huang

doi:10.1016/j.eswa.2025.129581

A deep reinforcement learning-guided multimodal multi-objective evolutionary algorithm with a serial-parallel mechanism

2025-09-19

0

PRE

AI

DOI：10.1016/j.eswa.2025.129581

Original

Original request for help

Save

Abstract

En 中文

• A novel DRL model is proposed for operator selection in MMOPs. • A neighborhood dominance-based ranking method is proposed to update the population. • A serial parallel mechanism is employed to enhance the diversity. • The proposed DRLMMEA outperforms 6 state-of-the-art multimodal multiobjective evolutionary algorithms on MMF test suites and a application named multimodal gearbox optimization problem.

A deep reinforcement learning-guided multimodal multi-objective evolutionary algorithm with a serial-parallel mechanism

Abstract

Journal

Organization