Travel the Same Path: A Novel TSP Solving Strategy

TL;DR

This paper introduces a novel TSP solving strategy using imitation learning and graph neural networks, achieving faster solutions for large instances while maintaining exactness and demonstrating strong generalization from small training instances.

Contribution

It presents a new imitation learning-based approach combined with graph neural networks for efficient and exact TSP solving, with proven generalization to larger instances.

Findings

Faster solution times for large TSP instances compared to baselines.

Maintains solution exactness despite speed improvements.

Strong generalization from small to large problem instances.

Abstract

In this paper, we provide a novel strategy for solving Traveling Salesman Problem, which is a famous combinatorial optimization problem studied intensely in the TCS community. In particular, we consider the imitation learning framework, which helps a deterministic algorithm making good choices whenever it needs to, resulting in a speed up while maintaining the exactness of the solution without suffering from the unpredictability and a potential large deviation. Furthermore, we demonstrate a strong generalization ability of a graph neural network trained under the imitation learning framework. Specifically, the model is capable of solving a large instance of TSP faster than the baseline while has only seen small TSP instances when training.