Continuously-Tempered PDMP Samplers

05/19/2022
by   Matthew Sutton, et al.
0

New sampling algorithms based on simulating continuous-time stochastic processes called piece-wise deterministic Markov processes (PDMPs) have shown considerable promise. However, these methods can struggle to sample from multi-modal or heavy-tailed distributions. We show how tempering ideas can improve the mixing of PDMPs in such cases. We introduce an extended distribution defined over the state of the posterior distribution and an inverse temperature, which interpolates between a tractable distribution when the inverse temperature is 0 and the posterior when the inverse temperature is 1. The marginal distribution of the inverse temperature is a mixture of a continuous distribution on [0,1) and a point mass at 1: which means that we obtain samples when the inverse temperature is 1, and these are draws from the posterior, but sampling algorithms will also explore distributions at lower temperatures which will improve mixing. We show how PDMPs, and particularly the Zig-Zag sampler, can be implemented to sample from such an extended distribution. The resulting algorithm is easy to implement and we show empirically that it can outperform existing PDMP-based samplers on challenging multimodal posteriors.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/24/2021

Annealed Leap-Point Sampler for Multimodal Target Distributions

In Bayesian statistics, exploring multimodal posterior distribution pose...
research
05/31/2019

Parallel Tempering via Simulated Tempering Without Normalizing Constants

In this paper we develop a new general Bayesian methodology that simulta...
research
12/11/2019

Sampling for Bayesian Mixture Models: MCMC with Polynomial-Time Mixing

We study the problem of sampling from the power posterior distribution i...
research
11/17/2015

Accelerating pseudo-marginal Metropolis-Hastings by correlating auxiliary variables

Pseudo-marginal Metropolis-Hastings (pmMH) is a powerful method for Baye...
research
10/13/2018

Optimal Temperature Spacing for Regionally Weight-preserving Tempering

Parallel tempering is popular method for allowing MCMC algorithms to pro...
research
03/07/2016

Partition Functions from Rao-Blackwellized Tempered Sampling

Partition functions of probability distributions are important quantitie...
research
06/12/2023

Generative Plug and Play: Posterior Sampling for Inverse Problems

Over the past decade, Plug-and-Play (PnP) has become a popular method fo...

Please sign up or login with your details

Forgot password? Click here to reset