Fetching the paper…

Reading the bibliography…

2019

Note on the bias and variance of variational inference

Huang, Chin-Wei, Courville, Aaron

Original
PDF
Post on X

Understand

In this note, we study the relationship between the variational gap and the variance of the (log) likelihood ratio.

We show that the gap can be upper bounded by some form of dispersion measure of the likelihood ratio, which suggests the bias of variational inference can be reduced by making the distribution of the likelihood ratio more concentrated, such as via averaging and variance reduction.

Note on the bias and variance of variational inference · Around

Built on

Importance weighted autoencoders
Yuri Burda, Roger Grosse, and Ruslan Salakhutdinov · 2015
Earlier work this paper cites.
Reinterpreting importance-weighted autoencoders
Original
Chris Cremer, Quaid Morris, and David Duvenaud · 2017
Earlier work this paper cites.
Filtering variational objectives
Chris J Maddison, John Lawson, George Tucker, Nicolas Heess, Mohammad Norouzi, Andriy Mnih, Arnaud Doucet, and Yee Teh · 2017
Earlier work this paper cites.

Similar

Importance weighting and variational inference
Justin Domke and Daniel R Sheldon · 2018
Cited alongside, same era.
Joint importance sampling for variational inference
Jack Klys, Jesse Bettencourt, and David Duvenaud · 2018
Cited alongside, same era.
Debiasing evidence approximations: On importance-weighted autoencoders and jackknife variational inference
Sebastian Nowozin · 2018
Cited alongside, same era.

Then

Hierarchical importance weighted autoencoders
Chin-Wei Huang, Kris Sankaran, Eeshan Dhekane, Alexandre Lacoste, and Aaron Courville · 2019
Closest in time.
Differentiable antithetic sampling for variance reduction in stochastic variational inference
Original
Mike Wu, Noah Goodman, and Stefano Ermon · 2019
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…