Graph Neural Network Guided Local Search for the Traveling Salesperson Problem
This work addresses the need for fast, high-quality solutions in real-time applications like transportation and logistics, presenting an incremental improvement over existing learning-based methods.
The paper tackles the challenge of solving large Traveling Salesperson Problem (TSP) instances quickly without sacrificing solution quality by proposing a hybrid approach using Graph Neural Networks (GNNs) and Guided Local Search (GLS). The results show a reduction in the mean optimality gap from 1.534% to 0.705% on 100-node problems and from 18.845% to 2.622% when generalizing from 20-node to 100-node instances.
Solutions to the Traveling Salesperson Problem (TSP) have practical applications to processes in transportation, logistics, and automation, yet must be computed with minimal delay to satisfy the real-time nature of the underlying tasks. However, solving large TSP instances quickly without sacrificing solution quality remains challenging for current approximate algorithms. To close this gap, we present a hybrid data-driven approach for solving the TSP based on Graph Neural Networks (GNNs) and Guided Local Search (GLS). Our model predicts the regret of including each edge of the problem graph in the solution; GLS uses these predictions in conjunction with the original problem graph to find solutions. Our experiments demonstrate that this approach converges to optimal solutions at a faster rate than three recent learning based approaches for the TSP. Notably, we reduce the mean optimality gap on the 100-node problem set from 1.534% to 0.705%, a 2x improvement. When generalizing from 20-node instances to the 100-node problem set, we reduce the optimality gap from 18.845% to 2.622%, a 7x improvement.