Abstract:Many Markov chain Monte Carlo techniques currently available rely on discrete-time reversible Markov processes whose transition kernels are variations of the Metropolis–Hastings algorithm. We explore and generalize an alternative scheme recently introduced in the physics literature (Peters and de With 2012) where the target distribution is explored using a continuous-time nonreversible piecewise-deterministic Markov process. In the Metropolis–Hastings algorithm, a trial move to a region of lower target density, equivalently of higher “energy,” than the current state can be rejected with positive probability. In this alternative approach, a particle moves along straight lines around the space and, when facing a high energy barrier, it is not rejected but its path is modified by bouncing against this barrier. By reformulating this algorithm using inhomogeneous Poisson processes, we exploit standard sampling techniques to simulate exactly this Markov process in a wide range of scenarios of interest. Additionally, when the target distribution is given by a product of factors dependent only on subsets of the state variables, such as the posterior distribution associated with a probabilistic graphical model, this method can be modified to take advantage of this structure by allowing computationally cheaper “local” bounces, which only involve the state variables associated with a factor, while the other state variables keep on evolving. In this context, by leveraging techniques from chemical kinetics, we propose several computationally efficient implementations. Experimentally, this new class of Markov chain Monte Carlo schemes compares favorably to state-of-the-art methods on various Bayesian inference tasks, including for high-dimensional models and large datasets. Supplementary materials for this article are available online.

Potential-Decomposition Strategy in Markov Chain Monte Carlo Sampling Algorithms

Enhancing Particle Swarm Optimization via probabilistic models

Efficient Irreversible Monte Carlo Samplers

A Splitting Hamiltonian Monte Carlo Method for Efficient Sampling

Stochastic Gradient Piecewise Deterministic Monte Carlo Samplers

The Bouncy Particle Sampler: A Nonreversible Rejection-Free Markov Chain Monte Carlo Method

A Random-Batch Monte Carlo Method for Many-Body Systems with Singular Kernels.

Efficient sampling using Macrocanonical Monte Carlo and density of states mapping

A Discrete Bouncy Particle Sampler

A Unified Particle-Optimization Framework for Scalable Bayesian Sampling

Direction-sweep Markov chains

Stratification and Optimal Resampling for Sequential Monte Carlo

Monte Carlo Approach to the Decay Rate of a Metastable System with an Arbitrarily Shaped Barrier

Superposition Enhanced Nested Sampling

Wide Sampling and Efficient Updating Monte Carlo Algorithms for Dimer Models

Dynamically Optimized Wang-Landau Sampling with Adaptive Trial Moves and Modification Factors

Metropolis-adjusted interacting particle sampling

Particle Optimization in Stochastic Gradient MCMC

Optimizing Few-Step Sampler for Diffusion Probabilistic Model

Efficient Sampling over Rough Energy Landscapes with High Barriers: A Combination of Metadynamics with Integrated Tempering Sampling.

Geometric allocation approaches in Markov chain Monte Carlo