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 state-of-the-art learning-based approaches and non-learning GLS algorithms for the TSP, notably finding optimal solutions to 96% of the 50-node problem set, 7% more than the next best benchmark, and to 20% of the 100-node problem set, 4.5x more than the next best benchmark. When generalizing from 20-node problems to the 100-node problem set, our approach finds solutions with an average optimality gap of 2.5%, a 10x improvement over the next best learning-based benchmark.
翻译:旅行销售者问题解决方案(TSP)在运输、物流和自动化流程中具有实际应用,然而,必须用最短的延迟来计算,以满足基本任务的实时性质。然而,在不牺牲解决方案质量的情况下快速解决大型TSP案例,对于目前的近似算法来说,仍然具有挑战性。为了缩小这一差距,我们提出了一个基于图表神经网络(GNN)和向导本地搜索(GLS)解决TSP的混合数据驱动方法。我们的模型预测了将问题图的每个边缘都纳入解决方案的遗憾;GLS利用这些预测与原始的问题图一起寻找解决方案。我们的实验表明,这一方法与最佳解决方案的趋同速度比最新学习方法和非学习的TSPGLS算法要快,特别是找到最佳解决方案,解决96%的50诺德问题,比下一个最佳基准多7%,以及100诺德问题设定的20%,比下一个最佳基准多4.5x。当从20诺德问题到下个100诺德问题图时,我们采用最佳解决方案,以平均2.5标准找到最佳解决方案。