MetFlow: A New Efficient Method for Bridging the Gap between Markov Chain Monte Carlo and Variational Inference

02/27/2020
by   Achille Thin, et al.
76

In this contribution, we propose a new computationally efficient method to combine Variational Inference (VI) with Markov Chain Monte Carlo (MCMC). This approach can be used with generic MCMC kernels, but is especially well suited to MetFlow, a novel family of MCMC algorithms we introduce, in which proposals are obtained using Normalizing Flows. The marginal distribution produced by such MCMC algorithms is a mixture of flow-based distributions, thus drastically increasing the expressivity of the variational family. Unlike previous methods following this direction, our approach is amenable to the reparametrization trick and does not rely on computationally expensive reverse kernels. Extensive numerical experiments show clear computational and performance improvements over state-of-the-art methods.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 7

page 8

page 20

page 21

page 22

page 23

page 24

page 25

03/23/2022

Robust Coordinate Ascent Variational Inference with Markov chain Monte Carlo simulations

Variational Inference (VI) is a method that approximates a difficult-to-...
08/04/2017

Learning Model Reparametrizations: Implicit Variational Inference by Fitting MCMC distributions

We introduce a new algorithm for approximate inference that combines rep...
11/17/2018

The Theory and Algorithm of Ergodic Inference

Approximate inference algorithm is one of the fundamental research field...
05/25/2018

Variational Measure Preserving Flows

Probabilistic modelling is a general and elegant framework to capture th...
02/17/2021

Divide-and-Conquer MCMC for Multivariate Binary Data

The analysis of large scale medical claims data has the potential to imp...
01/20/2021

Evaluating uncertainties in electrochemical impedance spectra of solid oxide fuel cells

Electrochemical impedance spectroscopy (EIS) is a widely used tool for c...
04/14/2016

Variational inference for rare variant detection in deep, heterogeneous next-generation sequencing data

The detection of rare variants is important for understanding the geneti...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.