投影式斯坦变分梯度下降

Feb, 2020

Projected Stein Variational Gradient Descent

Peng Chen, Omar Ghattas

TL;DR该研究提出一种基于投影斯坦变分梯度下降（pSVGD）方法，通过利用参数投影的低维系数，克服了贝叶斯推断中维度灾难的长期挑战，并在参数维度范围从数百到数万个的实验中证明了其准确性和效率。

Abstract

The curse of dimensionality is a critical challenge in bayesian inference for high dimensional parameters. In this work, we address this challenge by developing a →