HackerPapers

arXiv:1802.02538stat.ML

Yes, but Did It Work?: Evaluating Variational Inference

Yuling Yao, Aki Vehtari, Daniel Simpson, Andrew Gelman

Early discovery. Linked by one identifiable commenter. This paper is outside the main ranking; a mention is not an endorsement.

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 Pareto-smoothed importance sampling (PSIS) diagnostic gives a goodness of fit measurement for joint distributions, while simultaneously improving the error in the estimate. The variational simulation-based calibration (VSBC) assesses the average performance of point estimates.

arXiv comments
Appearing at International Conference on Machine Learning 2018
Journal reference
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:5581-5590, 2018. http://proceedings.mlr.press/v80/yao18a.html
DOI
Subjects
stat.MLMachine Learning (Statistics) stat.COComputation (Statistics)

On Hacker News

Linked by 1 identifiable commenter in 1 comment across 1 thread, from Feb 2023 to Feb 2023.

Comments linking this paper, per year.

Discussed in comments1 thread

  1. Hamiltonian Monte Carlo (2021) 1 comment

    • Amazing post, agree with everything you've said. I've always felt that the problems with advanced MCMC methods (HMC, RM-MC, etc) are even more painful when one looks at approximate bayesian methods - ADVI (variational approximation), SGLD (langevin dynamics), and so on. My grad…

      cfcf14View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.