Continuously-Tempered PDMP Samplers

no code implementations • 19 May 2022 • Matthew Sutton, Robert Salomone, Augustin Chevallier, Paul Fearnhead

We show how PDMPs, and particularly the Zig-Zag sampler, can be implemented to sample from such an extended distribution.

Efficient computation of the volume of a polytope in high-dimensions using Piecewise Deterministic Markov Processes

no code implementations • 18 Feb 2022 • Augustin Chevallier, Frédéric Cazals, Paul Fearnhead

Computing the volume of a polytope in high dimensions is computationally challenging but has wide applications.

Reversible Jump PDMP Samplers for Variable Selection

1 code implementation • 22 Oct 2020 • Augustin Chevallier, Paul Fearnhead, Matthew Sutton

A new class of Markov chain Monte Carlo (MCMC) algorithms, based on simulating piecewise deterministic Markov processes (PDMPs), have recently shown great promise: they are non-reversible, can mix better than standard MCMC algorithms, and can use subsampling ideas to speed up computation in big data scenarios.

Variable Selection