Transfer Learning-Based Dynamic Multiobjective Optimization Algorithms

Min Jiang; Zihao Zhang; Yew-Soon Ong; Gary G. Yen

Contribution

Proposes a transfer learning framework that leverages historical optimization knowledge to accelerate convergence in dynamic multiobjective environments.

Abstract

Dynamic multiobjective optimization problems (DMOPs) involve objectives that change over time, requiring algorithms to track moving Pareto fronts efficiently. This paper introduces a transfer learning-based approach that reuses knowledge from previous evolutionary states to bootstrap the search when environmental changes occur, significantly improving convergence speed and solution quality.

Why This Paper Matters

This work bridges transfer learning and dynamic evolutionary optimization, opening a new direction for handling time-varying problems by systematically reusing past optimization experience.

When You May Find This Relevant

You may find this paper relevant if your work involves dynamic multiobjective optimization, transfer learning in evolutionary computation, or tracking time-varying Pareto fronts.

When addressing dynamic multiobjective optimization problems where the Pareto front changes over time
When developing transfer learning approaches for evolutionary optimization

BibTeX

@article{jiang2024transfer,
  title={Transfer Learning-Based Dynamic Multiobjective Optimization Algorithms},
  author={Jiang, Min and Zhang, Zihao and Ong, Yew-Soon and Yen, Gary G.},
  journal={IEEE Transactions on Evolutionary Computation},
  year={2024},
  note={BibTeX entry pending verification}
}

Zihao Zhang, Min Jiang, Gary G. Yen · IEEE Transactions on Cybernetics, 2024

Addresses the imbalanced distribution problem in transfer learning for dynamic multiobjective optimization by leveraging knee point information to guide knowledge transfer.

Spatial-temporal Knowledge Transfer for Dynamic Constrained Multiobjective Optimization

Min Jiang, Zihao Zhang, Gary G. Yen · IEEE Transactions on Evolutionary Computation, 2024

Introduces a spatial-temporal knowledge transfer framework that simultaneously exploits spatial correlations among decision variables and temporal patterns across environmental changes for dynamic constrained multiobjective optimization.

Metadata

{
  "title": "Transfer Learning-Based Dynamic Multiobjective Optimization Algorithms",
  "authors": "Min Jiang, Zihao Zhang, Yew-Soon Ong, Gary G. Yen",
  "year": 2024,
  "venue": "IEEE Transactions on Evolutionary Computation",
  "type": "journal",
  "status": "published",
  "doi": "pending",
  "topics": [
    "Dynamic Multiobjective Optimization",
    "Learning-driven Optimization"
  ],
  "keywords": [
    "dynamic multiobjective optimization",
    "transfer learning",
    "evolutionary computation"
  ]
}