Less Is More - On the Importance of Sparsification for Transformers and Graph Neural Networks for TSP

arXiv:2403.17159v1 Announce Type: new
Abstract: Most of the recent studies tackling routing problems like the Traveling Salesman Problem (TSP) with machine learning use a transformer or Graph Neural Network (GNN) based encoder architecture. However, many of them apply these encoders naively by allowing them to aggregate information over the whole TSP instances. We, on the other hand, propose a data preprocessing method that allows the encoders to focus on the most relevant parts of the TSP instances only. In particular, …

abstract apply architecture arxiv cs.ai cs.lg encoder gnn graph graph neural network graph neural networks however importance machine machine learning network networks neural network neural networks routing studies them transformer transformers type