Provable algorithms for inference in topic models

Conference Paper

Recently, there has been considerable progress on designing algorithms with provable guarantees - typically using linear algebraic methods - for parameter learning in latent variable models. But designing provable algorithms for inference has proven to be more challenging. Here we take a first step towards provable inference in topic models. We leverage a property of topic models that enables us to construct simple linear estimators for the unknown topic proportions that have small variance, and consequently can work with short documents. Our estimators also correspond to finding an estimate around which the posterior is well-concentrated. We show lower bounds that for shorter documents it can be information theoretically impossible to find the hidden topics. Finally, we give empirical results that demonstrate that our algorithm works on realistic topic models. It yields good solutions on synthetic data and runs in time comparable to a single iteration of Gibbs sampling.

Duke Authors

Cited Authors

  • Arora, S; Ge, R; Koehler, F; Ma, T; Moitra, A

Published Date

  • January 1, 2016

Published In

  • 33rd International Conference on Machine Learning, Icml 2016

Volume / Issue

  • 6 /

Start / End Page

  • 4176 - 4184

International Standard Book Number 13 (ISBN-13)

  • 9781510829008

Citation Source

  • Scopus