Yes, but Did It Work?: Evaluating Variational Inference
[edit]
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:55815590, 2018.
Abstract
While it’s always possible to compute a variational approximation to a posterior distribution, it can be difficult to discover problems with this approximation. We propose two diagnostic algorithms to alleviate this problem. The Paretosmoothed importance sampling (PSIS) diagnostic gives a goodness of fit measurement for joint distributions, while simultaneously improving the error in the estimate. The variational simulationbased calibration (VSBC) assesses the average performance of point estimates.
Related Material


