Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 57 - Frontiers in Bayesian Computing
Type: Contributed
Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #313825
Title: Markov Chain Monte Carlo Methods with Sequential Proposals
Author(s): Joonha Park* and Yves Atchade
Companies: Boston University and Boston University
Keywords: Markov chain Monte Carlo; Hamiltonian Monte Carlo; The bouncy particle sampler; No-U-Turn sampler; Bayesian computation; Delayed rejection
Abstract:

We propose a novel Markov chain Monte Carlo (MCMC) approach, where candidates for the next state of the Markov chain are sequentially drawn. This sequential-proposal strategy can be applied to various existing MCMC methods, including Metropolis-Hastings algorithms and methods that use deterministic proposals such as Hamiltonian Monte Carlo (HMC) and the bouncy particle sampler. We introduce a unifying framework that defines a class of algorithms that include both the Hamiltonian Monte Carlo algorithm and the bouncy particle sampler, and explain how the sequential-proposal strategy can be combined for this class of algorithms. Using the sequential-proposal strategy, we also develop new Hamiltonian Monte Carlo methods that automatically tunes the lengths of the trajectories leading to proposals, similar to the widely used No-U-Turn sampler (NUTS). The numerical efficiencies of these new methods compare favorably to that of the NUTS. Finally, we demonstrate that the sequential-proposal strategy can enable the bouncy particle sampler to pass through regions of low target density and thus facilitate better mixing of the chain when the target is multimodal. (joint work with Yves Atchade)


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program