Dynamic multiobjective evolutionary algorithm based on a knee point driven Gaussian model

doi:10.1016/j.eswa.2025.129325

Dynamic multiobjective evolutionary algorithm based on a knee point driven Gaussian model

2025-08-19

0

PRE

AI

杨圣祥 (Shengxiang Yang)

Y

Yuan Liang

DOI：10.1016/j.eswa.2025.129325

Original

Original request for help

Save

Abstract

En 中文

• A knee point exploration strategy based on historical knee points and convergence related variables is designed to estimate the knee points under the new environment. • A generative model is constructed based on Gaussian distribution and diversity related variables, producing a high-quality initial population under the new environment. • The generative model is updated to generate offspring individuals in terms of the valuable knowledge extracted from the evolution, speeding up the convergence.

Keywords:

knee point exploration

generative model

Gaussian distribution

convergence speed

evolutionary algorithms

Dynamic multiobjective evolutionary algorithm based on a knee point driven Gaussian model

Abstract

Journal

Organization