README.md

Reinforcement Learning-based Non-Autoregressive Solver for Traveling Salesman Problems

The PyTorch Implementation of Arxiv-2308.00560 -- "Reinforcement Learning-based Non-Autoregressive Solver for Traveling Salesman Problems"pdf.

This manuscript has been accepted by TNNLS in Oct 2024, and the lastest version has been updated in Arxiv.

This paper propose the first non-autogressive model trained using reinforcement learning for solving TSPs.

How to Run

# 1. Training
python -u train.py

# 2. Testing
python -u val.py

Acknowledgments

• We would like to thank the following repository, which is the baseline of our code:

  • https://github.com/xbresson/TSP_Transformer

Citation

If you find our paper and code useful, please cite our paper (arxiv version):

@misc{Xiao2023,
      title={Reinforcement Learning-based Non-Autoregressive Solver for Traveling Salesman Problems}, 
      author={Yubin Xiao and Di Wang and Boyang Li and Huanhuan Chen and Wei Pang and Xuan Wu and Hao Li and Dong Xu and Yanchun Liang and You Zhou},
      year={2023},
      eprint={2308.00560},
      archivePrefix={arXiv},
}

Or the TNNLS version:

@ARTICLE{Xiao2024,
  author={Yubin Xiao and Di Wang and Boyang Li and Huanhuan Chen and Wei Pang and Xuan Wu and Hao Li and Dong Xu and Yanchun Liang and You Zhou},
  journal={IEEE Transactions on Neural Networks and Learning Systems}, 
  title={Reinforcement Learning-Based Nonautoregressive Solver for Traveling Salesman Problems}, 
  year={2024},
  volume={},
  number={},
  pages={1-15},
  doi={10.1109/TNNLS.2024.3483231}
}