Abstract:Recent systems applying Machine Learning (ML) to solve the Traveling Salesman Problem (TSP) exhibit issues when they try to scale up to real case scenarios with several hundred vertices. The use of Candidate Lists (CLs) has been brought up to cope with the issues. A CL is defined as a subset of all the edges linked to a given vertex such that it contains mainly edges that are believed to be found in the optimal tour. The initialization procedure that identifies a CL for each vertex in the TSP aids the solver by restricting the search space during solution creation. It results in a reduction of the computational burden as well, which is highly recommended when solving large TSPs. So far, ML was engaged to create CLs and values on the elements of these CLs by expressing ML preferences at solution insertion. Although promising, these systems do not restrict what the ML learns and does to create solutions, bringing with them some generalization issues. Therefore, motivated by exploratory and statistical studies of the CL behavior in multiple TSP solutions, in this work, we rethink the usage of ML by purposely employing this system just on a task that avoids well-known ML weaknesses, such as training in presence of frequent outliers and the detection of under-represented events. The task is to confirm inclusion in a solution just for edges that are most likely optimal. The CLs of the edge considered for inclusion are employed as an input of the neural network, and the ML is in charge of distinguishing when such edge is in the optimal solution from when it is not. The proposed approach enables a reasonable generalization and unveils an efficient balance between ML and optimization techniques. Our ML-Constructive heuristic is trained on small instances. Then, it is able to produce solutions—without losing quality—for large problems as well. We compare our method against classic constructive heuristics, showing that the new approach performs well for TSPLIB instances up to 1748 cities. Although ML-Constructive exhibits an expensive constant computation time due to training, we proved that the computational complexity in the worst-case scenario—for the solution construction after training—is O(n2logn2), n being the number of vertices in the TSP instance.

Improvements for mlrose applied to the Traveling Salesperson Problem

Optimization and Improvement of Genetic Algorithms Solving Traveling Salesman Problem

How to Evaluate Machine Learning Approaches for Combinatorial Optimization: Application to the Travelling Salesman Problem

A New Constructive Heuristic Driven by Machine Learning for the Traveling Salesman Problem

Two phase heuristic algorithm for the multiple-travelling salesman problem

Position: Rethinking Post-Hoc Search-Based Neural Approaches for Solving Large-Scale Traveling Salesman Problems

A Novel Genetic Algorithm Based on Circles for Larger-Scale Traveling Salesman Problem

Heuristics and Learning Models for Dubins MinMax Traveling Salesman Problem

Routing optimization with Monte Carlo Tree Search-based multi-agent reinforcement learning

An Empirical Analysis of Approximation Algorithms for the Euclidean Traveling Salesman Problem

An Investigation of Hybrid Tabu Search for the Traveling Salesman Problem.

A Hybrid Algorithm Based on Genetic Algorithm and Ant Colony Optimization for Traveling Salesman Problems

Local search for the Traveling Salesman Problem: A comparative study

New Heuristic Algorithms for Solving Single-Vehicle and Multi-Vehicle Generalized Traveling Salesman Problems (GTSP)

A Comparative Study of Improved GA and PSO in Solving Multiple Traveling Salesmen Problem

Bi-heuristic ant colony optimization-based approaches for traveling salesman problem

Optimization Approaches for the Traveling Salesman Problem with Drone

A Hybrid Cellular Genetic Algorithm for the Traveling Salesman Problem

Agent-based Self-organized Constructive Heuristics for Travelling Salesman Problem

Solving Traveling Salesman Problems Based on Artificial Cooperative Search Algorithm

Artificial rat optimization with decision-making: A bio-inspired metaheuristic algorithm for solving the traveling salesman problem