Sliced Wasserstein with Random-Path Projecting Directions

arXiv:2401.15889v2 Announce Type: replace-cross
Abstract: Slicing distribution selection has been used as an effective technique to improve the performance of parameter estimators based on minimizing sliced Wasserstein distance in applications. Previous works either utilize expensive optimization to select the slicing distribution or use slicing distributions that require expensive sampling methods. In this work, we propose an optimization-free slicing distribution that provides a fast sampling for the Monte Carlo estimation of expectation. In particular, we introduce the random-path projecting direction (RPD) …

abstract applications arxiv cs.ai cs.cv cs.lg distribution optimization path performance random sampling slicing stat.ml type work