Neural Combinatorial Optimization with Reinforcement Learning. Bello, I., Pham, H., Le, Q. V., Norouzi, M., & Bengio, S. In Workshop Track of the International Conference on Learning Representations, ICLR, 2017.
Neural Combinatorial Optimization with Reinforcement Learning [link]Paper  abstract   bibtex   
We present a framework to tackle combinatorial optimization problems using neural networks and reinforcement learning. We focus on the traveling salesman problem (TSP) and train a recurrent neural network that, given a set of city coordinates, predicts a distribution over different city permutations. Using negative tour length as the reward signal, we optimize the parameters of the recurrent neural network using a policy gradient method. Without much engineering and heuristic designing, Neural Combinatorial Optimization achieves close to optimal results on 2D Euclidean graphs with up to 100 nodes. These results, albeit still quite far from state-of-the-art, give insights into how neural networks can be used as a general tool for tackling combinatorial optimization problems.
@inproceedings{bello:2017:iclr_workshop,
  author = {I. Bello and H. Pham and Q. V. Le and M. Norouzi and S. Bengio},
  title = {Neural Combinatorial Optimization with Reinforcement Learning},
  booktitle = {Workshop Track of the International Conference on Learning Representations, {ICLR}},
  year = 2017,
  web = {https://openreview.net/forum?id=Bk9mxlSFx&noteId=Bk9mxlSFx},
  url = {publications/ps/bello_2017_iclr_workshop.ps.gz},
  pdf = {publications/pdf/bello_2017_iclr_workshop.pdf},
  djvu = {publications/djvu/bello_2017_iclr_workshop.djvu},
  abstract = {We present a framework to tackle combinatorial optimization problems using neural networks and reinforcement learning. We focus on the traveling salesman problem (TSP) and train a recurrent neural network that, given a set of city coordinates, predicts a distribution over different city permutations. Using negative tour length as the reward signal, we optimize the parameters of the recurrent neural network using a policy gradient method. Without much engineering and heuristic designing, Neural Combinatorial Optimization achieves close to optimal results on 2D Euclidean graphs with up to 100 nodes. These results, albeit still quite far from state-of-the-art, give insights into how neural networks can be used as a general tool for tackling combinatorial optimization problems.},
  categorie = {C},
}
Downloads: 0
About
Documentation
Keywords
Search
Users
Terms and Conditions of Use
Privacy Policy
Sitemap
© BibBase.org 2021