Machine Learning Constructives and Local Searches for the Travelling Salesman Problem

TL;DR

This paper enhances a hybrid machine learning and optimization heuristic for the Traveling Salesman Problem by reducing model complexity and integrating local search, leading to improved scalability and performance.

Contribution

It introduces computational weight reductions to the deep learning model and incorporates local search to enhance the ML-Constructive heuristic.

Findings

Improved model reduces execution time.

Local search enhances solution quality.

Experimental results confirm performance gains.

Abstract

The ML-Constructive heuristic is a recently presented method and the first hybrid method capable of scaling up to real scale traveling salesman problems. It combines machine learning techniques and classic optimization techniques. In this paper we present improvements to the computational weight of the original deep learning model. In addition, as simpler models reduce the execution time, the possibility of adding a local-search phase is explored to further improve performance. Experimental results corroborate the quality of the proposed improvements.