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.

The adaptive interpolation method for proving replica formulas. Applications to the Curie–Weiss and Wigner spike models

The adaptive interpolation method: a simple scheme to prove replica formulas in Bayesian inference

Replica symmetry breaking in neural networks: a few steps toward rigorous results

Strong replica symmetry in high-dimensional optimal Bayesian inference

Replica bound for Ising spin glass models in one dimension

Exact Fractional Inference via Re-Parametrization & Interpolation between Tree-Re-Weighted- and Belief Propagation- Algorithms

Generic proving of replica symmetry breaking

Fast Laplace Approximation for Sparse Bayesian Spike and Slab Models

Bayesian Interpolation

Replica-mean-field limits of fragmentation-interaction-aggregation processes

Bernstein -- von Mises theorems for statistical inverse problems II: Compound Poisson processes

Cavity Method: Message Passing from a Physics Perspective

Informed reversible jump algorithms

Information limits and Thouless-Anderson-Palmer equations for spiked matrix models with structured noise

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

No Statistical-Computational Gap in Spiked Matrix Models with Generative Network Priors

Replica-mean-field limit of continuous-time fragmentation-interaction-aggregation processes

A Non-Asymptotic Framework for Approximate Message Passing in Spiked Models

Solving the fully-connected spherical $p$-spin model with the cavity method: equivalence with the replica results

New Understanding of the Bethe Approximation and the Replica Method

A surrogate by exchangeability approach to the Curie-Weiss model