Implicit Variational Inference
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact jg801.
Variational inference is frequently the preferred method for modelling complicated posteriors arising from Bayesian inference. Classical variational methods restrict the approximate posterior to the exponential family, which may lead to large amounts of bias in the estimation of model parameters. Many methods have been devised in recent years for allowing more flexible posteriors. In this talk, we will discuss recent advances in variational inference with implicit distributions: distributions from which we can draw samples and compute gradients, but do not have analytic expression for.
Suggested reading:
Adversarial Variational Bayes – Mescheder et. al. 2017
Kernel Implicit Variational Inference – Shi et. al. 2018
Semi-Implicit Variational Inference – Yin & Zhou 2018
Unbiased Implicit Variational Inference – Titsias & Ruiz 2019
This talk is part of the Machine Learning Reading Group @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|