Test-Time Augmentation for Traveling Salesperson Problem

arXiv:2405.04767v1 Announce Type: new
Abstract: We propose Test-Time Augmentation (TTA) as an effective technique for addressing combinatorial optimization problems, including the Traveling Salesperson Problem. In general, deep learning models possessing the property of invariance, where the output is uniquely determined regardless of the node indices, have been proposed to learn graph structures efficiently. In contrast, we interpret the permutation of node indices, which exchanges the elements of the distance matrix, as a TTA scheme. The results demonstrate that our method …

abstract arxiv augmentation cs.lg deep learning general graph learn node optimization property test type