TY - GEN CY - Vancouver, Canada T3 - International Conference on Neural Information Processing Systems A1 - Korba, A A1 - Salim, A A1 - Arbel, M A1 - Luise, G A1 - Gretton, A PB - Neural Information Processing Systems Conference Y1 - 2020/12/06/ N2 - We study the Stein Variational Gradient Descent (SVGD) algorithm, which optimises a set of particles to approximate a target probability distribution ? ? e ?V on R d. In the population limit, SVGD performs gradient descent in the space of probability distributions on the KL divergence with respect to ?, where the gradient is smoothed through a kernel integral operator. In this paper, we provide a novel finite time analysis for the SVGD algorithm. We provide a descent lemma establishing that the algorithm decreases the objective at each iteration, and rates of convergence for the averaged Stein Fisher divergence (also referred to as Kernel Stein Discrepancy). We also provide a convergence result of the finite particle system corresponding to the practical implementation of SVGD to its population version. AV - public SN - 1049-5258 N1 - This version is the version of record. For information on re-use, please refer to the publisher?s terms and conditions. ID - discovery10166658 TI - A Non-Asymptotic Analysis for Stein Variational Gradient Descent UR - https://proceedings.neurips.cc/paper/2020/hash/3202111cf90e7c816a472aaceb72b0df-Abstract.html ER -