重要性加权和变分推断

Aug, 2018

Importance Weighting and Varational Inference

Justin Domke, Daniel Sheldon

TL;DR本文研究使用重要性采样方法对概率变分推断的影响，并提出了“重要性加权变分推断（IWVI）”技术，它是“增广变分推断”的一种实例，能够改善低维准确性和高维收敛性，实验证实了其对概率推断的实用性。

Abstract

Recent work used importance sampling ideas for better variational bounds on likelihoods. We clarify the applicability of these ideas to pure prob