Combining Reinforcement Learning and Optimal Transport for the Traveling Salesman Problem

Author name not available (Why is that?)

Publication date: 2 March 2022

Abstract: The traveling salesman problem is a fundamental combinatorial optimization problem with strong exact algorithms. However, as problems scale up, these exact algorithms fail to provide a solution in a reasonable time. To resolve this, current works look at utilizing deep learning to construct reasonable solutions. Such efforts have been very successful, but tend to be slow and compute intensive. This paper exemplifies the integration of entropic regularized optimal transport techniques as a layer in a deep reinforcement learning network. We show that we can construct a model capable of learning without supervision and inferences significantly faster than current autoregressive approaches. We also empirically evaluate the benefits of including optimal transport algorithms within deep learning models to enforce assignment constraints during end-to-end training.

Has companion code repository: https://github.com/badrmarani/tsp-attn

This page was built for publication: Combining Reinforcement Learning and Optimal Transport for the Traveling Salesman Problem

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6392554)