A double deep reinforcement learning-driven sparse large-scale multi-objective optimization algorithm

Jie Cao

doi:10.1016/j.asoc.2026.115214

A double deep reinforcement learning-driven sparse large-scale multi-objective optimization algorithm

2026-04-15

0

PRE

AI

DOI：10.1016/j.asoc.2026.115214

Original

Original request for help

Save

Abstract

En 中文

• An effective sparse large-scale multi-objective optimization algorithm is designed. • Applying two deep neural networks to sparse large-scale multi-objective evolutionary algorithms. • Design of a variable importance-guided mutation operator for generating high-quality sparse solutions. • The exploration of global regions and exploitation of local regions in decision space are enhanced.

Keywords:

sparse large-scale multi-objective optimization

deep neural networks

variable importance-guided mutation

global exploration

local exploitation

A double deep reinforcement learning-driven sparse large-scale multi-objective optimization algorithm

Abstract

Journal

Organization