The Bouncy Particle Sampler: A Non-Reversible Rejection-Free Markov Chain Monte Carlo Method
From MaRDI portal
Abstract: Markov chain Monte Carlo methods have become standard tools in statistics to sample from complex probability measures. Many available techniques rely on discrete-time reversible Markov chains whose transition kernels build up over the Metropolis-Hastings algorithm. We explore and propose several original extensions of an alternative approach introduced recently in Peters and de With (2012) where the target distribution of interest is explored using a continuous-time Markov process. In the Metropolis-Hastings algorithm, a trial move to a region of lower target density, equivalently "higher energy", than the current state can be rejected with positive probability. In this alternative approach, a particle moves along straight lines continuously around the space and, when facing a high energy barrier, it is not rejected but its path is modified by bouncing against this barrier. The resulting non-reversible Markov process provides a rejection-free MCMC sampling scheme. We propose several original techniques to simulate this continuous-time process exactly in a wide range of scenarios of interest to statisticians. When the target distribution factorizes as a product of factors involving only subsets of variables, such as the posterior distribution associated to a probabilistic graphical model, it is possible to modify the original algorithm to exploit this structure and update in parallel variables within each clique. We present several extensions by proposing methods to sample mixed discrete-continuous distributions and distributions restricted to a connected smooth domain. We also show that it is possible to move the particle using a general flow instead of straight lines. We demonstrate the efficiency of this methodology through simulations on a variety of applications and show that it can outperform Hybrid Monte Carlo schemes in interesting scenarios.
Recommendations
- Randomized Hamiltonian Monte Carlo as scaling limit of the bouncy particle sampler and dimension-free convergence rates
- A flexible particle Markov chain Monte Carlo method
- Particle Markov Chain Monte Carlo Methods
- Reversible Jump, Birth-and-Death and More General Continuous Time Markov Chain Monte Carlo Samplers
- Coordinate sampler: a non-reversible Gibbs-like MCMC sampler
- A dynamic programming approach to efficient sampling from Boltzmann distributions
- Analysis of high-dimensional continuous time Markov chains using the local bouncy particle sampler
- Irreversible Monte Carlo algorithms for efficient sampling
- Without-replacement sampling for particle methods on finite state spaces
- Efficient Construction of Reversible Jump Markov Chain Monte Carlo Proposal Distributions
Cites work
- Bayesian analysis of continuous time Markov chains with application to phylogenetic modelling
- Consistency and fluctuations for stochastic gradient Langevin dynamics
- Graphical models, exponential families, and variational inference
- scientific article; zbMATH DE number 425394 (Why is no real title available?)
- scientific article; zbMATH DE number 3954145 (Why is no real title available?)
- scientific article; zbMATH DE number 6781368 (Why is no real title available?)
- Markov chain Monte Carlo and irreversibility
- Monte Carlo strategies in scientific computing.
- Optimal scaling for various Metropolis-Hastings algorithms.
- Riemann manifold Langevin and Hamiltonian Monte Carlo methods. With discussion and authors' reply
- The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
Cited in
(only showing first 100 items - show all)- RZigZag
- Piecewise deterministic Markov processes for continuous-time Monte Carlo
- Piecewise deterministic Markov processes for scalable Monte Carlo on restricted domains
- Comment: A brief survey of the current state of play for Bayesian computation in data science at big-data scale
- Simulating Coulomb and log-gases with hybrid Monte Carlo algorithms
- Is there an analog of Nesterov acceleration for gradient-based MCMC?
- Subgeometric hypocoercivity for piecewise-deterministic Markov process Monte Carlo methods
- Accelerating numerical simulation of continuous-time Boolean satisfiability solver using discrete gradient
- Peskun-Tierney ordering for Markovian Monte Carlo: beyond the reversible scenario
- A piecewise deterministic Monte Carlo method for diffusion bridges
- Spatiotemporal blocking of the bouncy particle sampler for efficient inference in state-space models
- Markov chain simulation for multilevel Monte Carlo
- Randomized Hamiltonian Monte Carlo as scaling limit of the bouncy particle sampler and dimension-free convergence rates
- Large deviations for the empirical measure of the zig-zag process
- Piecewise deterministic Markov processes and their invariant measures
- Cores for piecewise-deterministic Markov processes used in Markov chain Monte Carlo
- Approximations of piecewise deterministic Markov processes and their convergence properties
- High-dimensional scaling limits of piecewise deterministic sampling algorithms
- Automatic zig-zag sampling in practice
- Sticky PDMP samplers for sparse and local inference problems
- Collective proposal distributions for nonlinear MCMC samplers: mean-field theory and fast implementation
- Sparse hard-disk packings and local Markov chains
- On explicit \(L^2\)-convergence rate estimate for piecewise deterministic Markov processes in MCMC algorithms
- Statistical challenges in tracking the evolution of SARS-CoV-2
- Complexity of zigzag sampling algorithm for strongly log-concave distributions
- Spectral analysis of the zigzag process
- Reversible and non-reversible Markov chain Monte Carlo algorithms for reservoir simulation problems
- Optimal scaling of the MALA algorithm with irreversible proposals for Gaussian targets
- Markov chain Monte Carlo algorithms with sequential proposals
- Low-lying eigenvalues and convergence to the equilibrium of some piecewise deterministic Markov processes generators in the small temperature regime
- On the convergence time of some non-reversible Markov chain Monte Carlo methods
- Large-scale inference of correlation among mixed-type biological traits with phylogenetic multivariate probit models
- Regeneration-enriched Markov processes with application to Monte Carlo
- Hypocoercivity of piecewise deterministic Markov process-Monte Carlo
- Modified Hamiltonian Monte Carlo for Bayesian inference
- Coordinate sampler: a non-reversible Gibbs-like MCMC sampler
- Subsampling MCMC -- an introduction for the survey statistician
- Irreversible samplers from jump and continuous Markov processes
- Bayesian nonparametric clustering for large data sets
- Ergodicity of the zigzag process
- Geometric ergodicity of the bouncy particle sampler
- Adaptive schemes for piecewise deterministic Monte Carlo algorithms
- PDMP characterisation of event-chain Monte Carlo algorithms for particle systems
- Stochastic gradient Langevin dynamics with adaptive drifts
- Analysis of high-dimensional continuous time Markov chains using the local bouncy particle sampler
- scientific article; zbMATH DE number 7387626 (Why is no real title available?)
- Connecting the Dots: Numerical Randomized Hamiltonian Monte Carlo with State-Dependent Event Rates
- The Block-Poisson Estimator for Optimally Tuned Exact Subsampling MCMC
- Forward Event-Chain Monte Carlo: Fast Sampling by Randomness Control in Irreversible Markov Chains
- Kinetic walks for sampling
- Transport map accelerated adaptive importance sampling, and application to inverse problems arising from multiscale stochastic reaction networks
- Constructing sampling schemes via coupling: Markov semigroups and optimal transport
- Limit theorems for the zig-zag process
- Generalizing Parallel Replica Dynamics: Trajectory Fragments, Asynchronous Computing, and PDMPs
- Stochastic gradient Markov chain Monte Carlo
- Birth–death dynamics for sampling: global convergence, approximations and their asymptotics
- Infinite dimensional piecewise deterministic Markov processes
- Advanced Multilevel Monte Carlo Methods
- Transport Monte Carlo: High-Accuracy Posterior Approximation via Random Transport
- Efficient Simulation of Sparse Graphs of Point Processes
- Gradient-Based Markov Chain Monte Carlo for Bayesian Inference With Non-differentiable Priors
- Non-reversible guided Metropolis kernel
- Adaptive Importance Sampling Based on Fault Tree Analysis for Piecewise Deterministic Markov Process
- Speed up Zig-Zag
- On the accept-reject mechanism for Metropolis-Hastings algorithms
- A fresh Take on ‘Barker Dynamics’ for MCMC
- Polynomial convergence rates of piecewise deterministic Markov processes
- Hamiltonian-Assisted Metropolis Sampling
- Concave-Convex PDMP-based Sampling
- Reversible Jump PDMP Samplers for Variable Selection
- Randomized time Riemannian manifold Hamiltonian Monte Carlo
- Strong invariance principles for ergodic Markov processes
- Sampling algorithms in statistical physics: a guide for statistics and machine learning
- A langevinized ensemble Kalman filter for large-scale dynamic learning
- Speeding up the zig-zag process
- Sampling using adaptive regenerative processes
- Posterior computation with the Gibbs zig-zag sampler
- Contraction rate estimates of stochastic gradient kinetic Langevin integrators
- Non-reversible lifts of reversible diffusion processes and relaxation times
- Hypocoercivity meets lifts
- Perturbation theory for killed Markov processes and quasi-stationary distributions
- The velocity jump Langevin process and its splitting scheme: long time convergence and numerical accuracy
- Debiasing piecewise deterministic Markov process samplers using couplings
- Hoeffding's inequality for continuous-time Markov chains via the spectral gap
- Eyring-Kramers type formulas for some piecewise deterministic Markov processes
- Zigzag Path Connects Two Monte Carlo Samplers: Hamiltonian Counterpart to a Piecewise Deterministic Markov Process
- Scaling of piecewise deterministic Monte Carlo for anisotropic targets
- Optimal scaling results for Moreau-Yosida Metropolis-adjusted Langevin algorithms
- Bayesian mechanics for stationary processes
- HMC and underdamped Langevin united in the unadjusted convex smooth case
- Numerical approximations and convergence analysis of piecewise diffusion Markov processes, with application to glioma cell migration
- PDMP Monte Carlo methods for piecewise smooth densities
- Stereographic Markov chain Monte Carlo
- Controlled stochastic processes for simulated annealing
- Fused \(L_{1/2}\) prior for large scale linear inverse problem with Gibbs bouncy particle sampler
- Consistent multiscale modelling of movement and habitat selection
- A tractable symmetric multivariate logistic distribution with bounded score function
- Windowed thinning and query complexity for the bouncy particle and Zigzag samplers
- Theoretical guarantees for lifted samplers
- Piecewise deterministic sampling with splitting schemes
This page was built for publication: The Bouncy Particle Sampler: A Non-Reversible Rejection-Free Markov Chain Monte Carlo Method
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q144293)
