Learn-Improvement-Heuristics-for-Routing

A deep reinforcement learning framework to learn the improvement heuristics （with pairwise local operators, e.g. 2-opt, swap, reinsertion） for routing problems.

Paper

For more details, please refer to: Yaoxin Wu, Wen Song, Zhiguang Cao, Jie Zhang and Andrew Lim. Learning Improvement Heuristics for Solving Routing Problems, IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 9, pp. 5057-5069, 2022 (https://ieeexplore.ieee.org/document/9393606).

@article{wu2021learning,
  title={Learning Improvement Heuristics for Solving Routing Problems},
  author={Wu, Yaoxin and Song, Wen and Cao, Zhiguang and Zhang, Jie and Lim, Andrew},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2022},
  volume={33},
  number={9},
  pages={5057-5069}
}

Running the program

For each problem and size:

Training

python3 run.py

or run jupyter file

RUN MY MODEL.ipynb

Testing

python3 test.py

or run jupyter file

Test with model.ipynb

Visualization of Solving TSP and CVRP

https://youtu.be/97ZXp9zSEK8