Variational Boosting: Iteratively Refining Posterior Approximations
Andrew Miller · Nicholas J Foti · Ryan P. Adams

Tue Aug 8th 06:30 -- 10:00 PM @ Gallery #109

We propose a black-box variational inference method to approximate intractable distributions with an increasingly rich approximating class. Our method, variational boosting, iteratively refines an existing variational approximation by solving a sequence of optimization problems, allowing a trade-off between computation time and accuracy. We 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 the resulting posterior inferences compare favorably to existing variational algorithms.

Author Information

Andrew Miller (Harvard)
Nick J Foti (University of Washington)
Ryan P. Adams (Google Brain and Princeton University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors