Techniques for proving Asynchronous Convergence results for Markov Chain Monte Carlo methods

11/17/2017
by   Alexander Terenin, et al.
0

Markov Chain Monte Carlo (MCMC) methods such as Gibbs sampling are finding widespread use in applied statistics and machine learning. These often lead to difficult computational problems, which are increasingly being solved on parallel and distributed systems such as compute clusters. Recent work has proposed running iterative algorithms such as gradient descent and MCMC in parallel asynchronously for increased performance, with good empirical results in certain problems. Unfortunately, for MCMC this parallelization technique requires new convergence theory, as it has been explicitly demonstrated to lead to divergence on some examples. Recent theory on Asynchronous Gibbs sampling describes why these algorithms can fail, and provides a way to alter them to make them converge. In this article, we describe how to apply this theory in a generic setting, to understand the asynchronous behavior of any MCMC algorithm, including those implemented using parameter servers, and those not based on Gibbs sampling.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/24/2022

Understanding Linchpin Variables in Markov Chain Monte Carlo

An introduction to the use of linchpin variables in Markov chain Monte...
research
08/06/2020

Gibbs Sampling with People

A core problem in cognitive science and machine learning is to understan...
research
08/17/2019

Prune Sampling: a MCMC inference technique for discrete and deterministic Bayesian networks

We introduce and characterise the performance of the Markov chain Monte ...
research
02/04/2019

Is There an Analog of Nesterov Acceleration for MCMC?

We formulate gradient-based Markov chain Monte Carlo (MCMC) sampling as ...
research
08/22/2019

Gibbs sampling for game-theoretic modeling of private network upgrades with distributed generation

Renewable energy is increasingly being curtailed, due to oversupply or n...
research
08/18/2019

StreamNet: A DAG System with Streaming Graph Computing

To achieve high throughput in the POW based blockchain systems, a series...
research
12/02/2016

Asynchronous Stochastic Gradient MCMC with Elastic Coupling

We consider parallel asynchronous Markov Chain Monte Carlo (MCMC) sampli...

Please sign up or login with your details

Forgot password? Click here to reset