Variational Predictive Information Bottleneck
Machine Learning
2019-10-25 v1 Information Theory
math.IT
Machine Learning
Abstract
In classic papers, Zellner demonstrated that Bayesian inference could be derived as the solution to an information theoretic functional. Below we derive a generalized form of this functional as a variational lower bound of a predictive information bottleneck objective. This generalized functional encompasses most modern inference procedures and suggests novel ones.
Cite
@article{arxiv.1910.10831,
title = {Variational Predictive Information Bottleneck},
author = {Alexander A. Alemi},
journal= {arXiv preprint arXiv:1910.10831},
year = {2019}
}