• Home
  • Search
  • Diffusion Learning-guided Evolution for Large-scale Dynamic Multi-Objective Optimization
  • https://doi.org/10.1109/tevc.2026.3659883Copy DOI Icon

Diffusion Learning-guided Evolution for Large-scale Dynamic Multi-Objective Optimization

Show More
  • Abstract
  • Literature Map
  • Similar Papers
Abstract

Large-scale, dynamic multi-objective optimization problems (LSDMOPs) extend traditional DMOPs into high-dimensional decision spaces, reflecting the growing complexity of real-world dynamic systems. However, the effectiveness of existing dynamic multi-objective evolutionary algorithms is severely limited for LSDMOPs, due to inadequate training data, predictions in unknown environments, and large-scale dynamic search spaces. To address these challenges, we propose a diffusion learning-based evolutionary framework, inspired by the intrinsic analogy between iterative evolution of optimization search and stepwise denoising of diffusion model. Firstly, a new training paradigm is designed to treat evolutionary trajectories of population across environments as a learnable supervised source, thereby constructing rich training data to learn the changing patterns of optimal regions in dynamic fitness landscapes. Secondly, we introduce a trajectory alignment loss which encourages the stepwise denoising process to conform to the true population evolutionary behaviors in terms of spatial exploration, convergence trends, and boundary adaptation. Thirdly, a conditional diffusion model-based evolutionary strategy is proposed, which can gradually control denoising direction and intensity using predefined conditions, thereby generating optimization paths from noise toward Pareto-optimal solutions in a new environment. Experimental results demonstrate that the proposed framework outperforms state-of-the-art designs on a typical dynamic multi-objective test suite and real-world large-scale dynamic scenarios.

Similar Papers
  • Components

Supp1-3135020.pdf

  • Dec 24, 2021
  • Liang Feng
  • Research Article
  • Citations125

Solving Dynamic Multiobjective Problem via Autoencoding Evolutionary Search.

  • Oct 01, 2020
  • IEEE Transactions on Cybernetics
  • Liang Feng +4
  • Conference Article

Solving Dynamic Multi-Objective Optimization Problems Using Cultural Algorithm based on Decomposition

  • Aug 26, 2019
  • Ramya Ravichandran +1
  • Research Article
  • Citations68

Knowledge guided Bayesian classification for dynamic multi-objective optimization

  • Jun 04, 2022
  • Knowledge-Based Systems
  • Yulong Ye +5
  • Research Article
  • Citations60

Reducing Negative Transfer Learning via Clustering for Dynamic Multiobjective Optimization

  • Oct 01, 2022
  • IEEE Transactions on Evolutionary Computation
  • Jianqiang Li +4
  • Conference Article
  • Citations15

New Dynamic Multiobjective Evolutionary Algorithm with Core Estimation of Distribution

  • Jun 01, 2010
  • Chun-An Liu
  • PDF
  • Research Article
  • Citations1

Dynamic Multiobjective Optimization with Multiple Response Strategies Based on Linear Environment Detection

  • Nov 24, 2020
  • Complexity
  • Qiyuan Yu +4
  • Research Article
  • Citations42

Combining a hybrid prediction strategy and a mutation strategy for dynamic multiobjective optimization

  • Apr 01, 2022
  • Swarm and Evolutionary Computation
  • Ying Chen +5
  • Research Article
  • Citations52

Cooperative particle swarm optimization with reference-point-based prediction strategy for dynamic multiobjective optimization

  • Dec 03, 2019
  • Applied Soft Computing
  • Xiao-Fang Liu +2
  • Research Article
  • Citations92

A new prediction strategy for dynamic multi-objective optimization using Gaussian Mixture Model

  • Aug 22, 2021
  • Information Sciences
  • Feng Wang +3
  • Conference Article

The Effect of Quantum and Charged Particles on the Performance of the Dynamic Vector-evaluated Particle Swarm Optimisation Algorithm

  • Jul 11, 2015
  • Mardé Helbig +1
  • Conference Article
  • Citations6

Multi-objective optimization in dynamic environment: A review

  • Aug 01, 2011
  • Rui Chen +1
  • PDF
  • Research Article
  • Citations9

Transfer Learning Based on Clustering Difference for Dynamic Multi-Objective Optimization

  • Apr 11, 2023
  • Applied Sciences
  • Fangpei Yao +1
  • Research Article
  • Citations36

A pareto-based evolutionary algorithm using decomposition and truncation for dynamic multi-objective optimization

  • Aug 21, 2019
  • Applied Soft Computing
  • Junwei Ou +7
  • Conference Article
  • Citations1

Using heterogeneous knowledge sharing strategies with dynamic vector-evaluated particle swarm optimisation

  • Dec 01, 2014
  • Marde Helbig +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.