Regularized Stein Variational Gradient Flow

arXiv:2211.07861v2 Announce Type: replace-cross
Abstract: The Stein Variational Gradient Descent (SVGD) algorithm is a deterministic particle method for sampling. However, a mean-field analysis reveals that the gradient flow corresponding to the SVGD algorithm (i.e., the Stein Variational Gradient Flow) only provides a constant-order approximation to the Wasserstein Gradient Flow corresponding to the KL-divergence minimization. In this work, we propose the Regularized Stein Variational Gradient Flow, which interpolates between the Stein Variational Gradient Flow and the Wasserstein Gradient Flow. We establish …

abstract algorithm analysis approximation arxiv cs.lg cs.na divergence flow gradient however kl-divergence math.ap math.na math.st mean particle sampling stat.co stat.ml stat.th type