Variationally inferred sampling through a refined bound

Impacto

Downloads

Downloads per month over past year

Gallego, Víctor and Rios Insua, David (2021) Variationally inferred sampling through a refined bound. Entropy, 23 (1). p. 123. ISSN 1099-4300

[thumbnail of gallego_variationally_entropy.pdf]
Preview
PDF
1MB

Official URL: https://doi.org/10.3390/e23010123



Abstract

In this work, a framework to boost the efficiency of Bayesian inference in probabilistic models is introduced by embedding a Markov chain sampler within a variational posterior approximation. We call this framework “refined variational approximation”. Its strengths are its ease of implementation and the automatic tuning of sampler parameters, leading to a faster mixing time through automatic differentiation. Several strategies to approximate evidence lower bound (ELBO) computation are also introduced. Its efficient performance is showcased experimentally using statespace models for time-series data, a variational encoder for density estimation and a conditional variational autoencoder as a deep Bayes classifier.


Item Type:Article
Uncontrolled Keywords:Variational inference, MCMC, Stochastic gradients, Neural networks
Subjects:Sciences > Mathematics > Probabilities
ID Code:75595
Deposited On:17 Nov 2022 11:19
Last Modified:18 Nov 2022 08:16

Origin of downloads

Repository Staff Only: item control page