arxivst stuff from arxiv that you should probably bookmark

Sticking the Landing: An Asymptotically Zero-Variance Gradient Estimator for Variational Inference

Abstract · Mar 27, 2017 17:25 ·

stat-ml cs-lg

Arxiv Abstract

  • Geoffrey Roeder
  • Yuhuai Wu
  • David Duvenaud

We propose a simple and general variant of the standard reparameterized gradient estimator for the variational evidence lower bound. Specifically, we remove a part of the total derivative with respect to the variational parameters that corresponds to the score function. Removing this term produces an unbiased gradient estimator whose variance approaches zero as the approximate posterior approaches the exact posterior. We analyze the behavior of this gradient estimator theoretically and empirically, and generalize it to more complex variational distributions such as mixtures and importance-weighted posteriors.

Read the paper (pdf) »