Hybrid Pointer Networks for Traveling Salesman Problems Optimization

TL;DR

This paper introduces a hybrid pointer network that improves solving the Traveling Salesman Problem by combining different encoding methods, outperforming previous models in solution quality without relying on traditional optimization heuristics.

Contribution

The paper presents a novel hybrid pointer network architecture that enhances graph pointer networks for TSP, achieving better results without traditional heuristics.

Findings

Outperforms graph pointer networks on TSP50

Achieves lower tour length (5.706 vs 5.959)

Comparable to highly tuned algorithms

Abstract

In this work, a novel idea is presented for combinatorial optimization problems, a hybrid network, which results in a superior outcome. We applied this method to graph pointer networks [1], expanding its capabilities to a higher level. We proposed a hybrid pointer network (HPN) to solve the travelling salesman problem trained by reinforcement learning. Furthermore, HPN builds upon graph pointer networks which is an extension of pointer networks with an additional graph embedding layer. HPN outperforms the graph pointer network in solution quality due to the hybrid encoder, which provides our model with a verity encoding type, allowing our model to converge to a better policy. Our network significantly outperforms the original graph pointer network for small and large-scale problems increasing its performance for TSP50 from 5.959 to 5.706 without utilizing 2opt, Pointer networks, Attention model, and a wide range of models, producing results comparable to highly tuned and specialized algorithms. We make our data, models, and code publicly available [2].