Variational Prediction

Alexander A. Alemi, Ben Poole
7/14/2023
cs.LGstat.ML

No official code found. Best implementation: cnexah/VA-DepthNet (112 stars, MIT, last commit 2023-04-28).

Abstract

Bayesian inference offers benefits over maximum likelihood, but it also comes with computational costs. Computing the posterior is typically intractable, as is marginalizing that posterior to form the posterior predictive distribution. In this paper, we present variational prediction, a technique for directly learning a variational approximation to the posterior predictive distribution using a variational bound. This approach can provide good predictive distributions without test time marginalization costs. We demonstrate Variational Prediction on an illustrative toy example.

View on arXivView PDF

Code Implementations(1)

cnexah/VA-DepthNetCommunity77%
MIT

VA-DepthNet: A Variational Approach to Single Image Depth Prediction

1125PythonFeb 11, 20233 years agoMIT

Similar Papers

Predictive variational inference: Learn the predictively optimal posterior distribution

91% match2024

A Tutorial on Parametric Variational Inference

85% match2023

Variational Bayes Made Easy

83% match2023

Machine Learning and the Future of Bayesian Computation

82% match2023

Variational Inference with Coverage Guarantees in Simulation-Based Inference

82% match1 repo12023

Logistic Variational Bayes Revisited

81% match1 repo22024

Cite this paper

@article{alemi2023variational,
  title  = {Variational Prediction},
  author = {Alexander A. Alemi and Ben Poole},
  year   = {2023},
  eprint = {2307.07568},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2307.07568v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2307.07568.svg)](https://codeofpaper.com/paper/2307.07568)

Discussion