We propose a black-box variational inference method to approximate intractable distributions with an increasingly rich approximating class. Our method, termed variational boosting, iteratively refines an existing variational approximation by solving a sequence of optimization problems, allowing the practitioner to trade computation time for accuracy. We show how to expand the variational approximating class by incorporating additional covariance structure and by introducing new components to form a mixture. We apply variational boosting to synthetic and real statistical models, and show that resulting posterior inferences compare favorably to existing posterior approximation algorithms in both accuracy and efficiency.
@conference{miller2017boosting,
year = {2017},
author = {Miller, Andrew C. and Foti, Nicholas J. and Adams, Ryan P.},
title = {Variational Boosting: Iteratively Refining Posterior Approximations},
booktitle = {Proceedings of the 34th International Conference on Machine Learning (ICML)},
keywords = {ICML, variational inference, Bayesian methods},
note = {arXiv:1611.06585 [stat.ML]}
}