Approximation Based Variance Reduction for Reparameterization Gradients

Part of Advances in Neural Information Processing Systems 33 (NeurIPS 2020)

AuthorFeedback Bibtex MetaReview Paper Review Supplemental


Tomas Geffner, Justin Domke


Flexible variational distributions improve variational inference but are harder to optimize. In this work we present a control variate that is applicable for any reparameterizable distribution with known mean and covariance, e.g. Gaussians with any covariance structure. The control variate is based on a quadratic approximation of the model, and its parameters are set using a double-descent scheme. We empirically show that this control variate leads to large improvements in gradient variance and optimization convergence for inference with non-factorized variational distributions.