EPSL

Code for TEVC2025 Paper: Dealing with Structure Constraints in Evolutionary Pareto Set Learning

The code is mainly designed to be simple and readable, it contains:

• run_[epsl].py is a ~300-line main file to run the EPSL method, calculate the hypervolume, and plot the Pareto front;
• model.py or model_[shared_component/shared_component_syn/variable_relation/keypoint].py contains the standard EPSL model or EPSL model with structure constraint of [shared_component/shared_component_syn/variable_relation/keypoint];
• problem.py contains all test problems used in this paper;
• The folder data contains the problem information for the RE problems, which is obtained from the reproblems repository.

Reference

If you find our work helpful for your research, please cite our paper:

@article{lin2025dealing,
  title={Dealing With Structure Constraints in Evolutionary Pareto Set Learning},
  author={Lin, Xi and Zhang, Xiaoyuan and Yang, Zhiyuan and Zhang, Qingfu},
  journal={IEEE Transactions on Evolutionary Computation},
  year={2025},
  publisher={IEEE}
}

If you find the RE problems useful for your research, please also cite the RE paper:

@article{tanabe2020easy,
  title={An easy-to-use real-world multi-objective optimization problem suite},
  author={Tanabe, Ryoji and Ishibuchi, Hisao},
  journal={Applied Soft Computing},
  volume={89},
  pages={106078},
  year={2020},
  publisher={Elsevier}
}