BriefGPT.xyz
Aug, 2017
提升变分推断：优化视角
Boosting Variational Inference: an Optimization Perspective
HTML
PDF
Francesco Locatello, Rajiv Khanna, Joydeep Ghosh, Gunnar Rätsch
TL;DR
本文研究了基于 boosting 的变分推断算法在优化视角下的收敛性质及其与经典 Frank-Wolfe 算法之间的联系，提出了收敛的充分条件，明确了算法的收敛速率和简化方法，并在概率模型和理论特性之间架起了桥梁。
Abstract
variational inference
is a popular technique to approximate a possibly intractable
bayesian posterior
with a more tractable one. Recently,
boosti
→