Stochastic Gradient Riemannian Langevin Dynamics on the Probability Simplex

Part of Advances in Neural Information Processing Systems 26 (NIPS 2013)

Bibtex Metadata Paper Reviews Supplemental

Authors

Sam Patterson, Yee Whye Teh

Abstract

In this paper we investigate the use of Langevin Monte Carlo methods on the probability simplex and propose a new method, Stochastic gradient Riemannian Langevin dynamics, which is simple to implement and can be applied online. We apply this method to latent Dirichlet allocation in an online setting, and demonstrate that it achieves substantial performance improvements to the state of the art online variational Bayesian methods.