Combining Reinforcement Learning with Lin-Kernighan-Helsgaun Algorithm for the Traveling Salesman Problem
Abstract
We address the Traveling Salesman Problem (TSP), a famous NP-hard combinatorial optimization problem. And we propose a variable strategy reinforced approach, denoted as VSR-LKH, which combines three reinforcement learning methods (Q-learning, Sarsa and Monte Carlo) with the well-known TSP algorithm, called Lin-Kernighan-Helsgaun (LKH). VSR-LKH replaces the inflexible traversal operation in LKH, and lets the program learn to make choice at each search step by reinforcement learning. Experimental results on 111 TSP benchmarks from the TSPLIB with up to 85,900 cities demonstrate the excellent performance of the proposed method.
Cite
Text
Zheng et al. "Combining Reinforcement Learning with Lin-Kernighan-Helsgaun Algorithm for the Traveling Salesman Problem." AAAI Conference on Artificial Intelligence, 2021. doi:10.1609/AAAI.V35I14.17476Markdown
[Zheng et al. "Combining Reinforcement Learning with Lin-Kernighan-Helsgaun Algorithm for the Traveling Salesman Problem." AAAI Conference on Artificial Intelligence, 2021.](https://mlanthology.org/aaai/2021/zheng2021aaai-combining/) doi:10.1609/AAAI.V35I14.17476BibTeX
@inproceedings{zheng2021aaai-combining,
title = {{Combining Reinforcement Learning with Lin-Kernighan-Helsgaun Algorithm for the Traveling Salesman Problem}},
author = {Zheng, Jiongzhi and He, Kun and Zhou, Jianrong and Jin, Yan and Li, Chu-Min},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2021},
pages = {12445-12452},
doi = {10.1609/AAAI.V35I14.17476},
url = {https://mlanthology.org/aaai/2021/zheng2021aaai-combining/}
}