Piecewise Deterministic Markov Processes for Scalable Monte Carlo on Restricted Domains

Joris Bierkens, Alexandre Bouchard-Côté, Arnaud Doucet, Andrew B. Duncan, Paul Fearnhead +3 more
1/16/2017
stat.MEstat.CO

Official code: alan-turing-institute/PDSampler.jl (33 stars, last commit 2020-11-29).

Abstract

Piecewise Deterministic Monte Carlo algorithms enable simulation from a posterior distribution, whilst only needing to access a sub-sample of data at each iteration. We show how they can be implemented in settings where the parameters live on a restricted domain.

View on arXivView PDF

Code Implementations(1)

alan-turing-institute/PDSampler.jlOfficial100%

Piecewise Deterministic Sampler library (Bouncy particle sampler, Zig Zag sampler, ...)

338JuliaApr 6, 20175 years agoNOASSERTION
bayesian-inferencebouncy-particle-samplergraphical-modelsmarkov-chain-monte-carlosampling-methods+1 more

Similar Papers

Asynchronous, Generation-Free ABC: Streaming AMIS Reweighting for Heterogeneous Simulator Workloads on HPC

84% match2026

The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data

80% match1 repo72016

Amortized Bayesian Decision Making for simulation-based models

79% match1 repo32023

Combining Monte Carlo Tree Search and Heuristic Search for Weighted Vertex Coloring

78% match2023

Accelerating Approximate Thompson Sampling with Underdamped Langevin Monte Carlo

78% match1 repo42024

Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows

78% match1 repo432018

Cite this paper

@article{bierkens2017piecewise,
  title  = {Piecewise Deterministic Markov Processes for Scalable Monte Carlo on Restricted Domains},
  author = {Joris Bierkens and Alexandre Bouchard-Côté and Arnaud Doucet and Andrew B. Duncan and Paul Fearnhead and Thibaut Lienart and Gareth Roberts and Sebastian J. Vollmer},
  year   = {2017},
  eprint = {1701.04244},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1701.04244}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1701.04244.svg)](https://codeofpaper.com/paper/1701.04244)

Discussion