Zig-zag sampling for discrete structures and non-reversible phylogenetic MCMC
arXiv:2004.08807 · doi:10.1080/10618600.2022.2032722
Abstract
We construct a zig-zag process targeting a posterior distribution defined on a hybrid state space consisting of both discrete and continuous variables. The construction does not require any assumptions on the structure among discrete variables. We demonstrate our method on two examples in genetics based on the Kingman coalescent, showing that the zig-zag process can lead to efficiency gains of up to several orders of magnitude over classical Metropolis-Hastings algorithms, and that it is well suited to parallel computation. Our construction resembles existing techniques for Hamiltonian Monte Carlo on a hybrid state space, which suffers from implementationally and analytically complex boundary crossings when applied to the coalescent. We demonstrate that the continuous-time zig-zag process avoids these complications.
arXiv comment22 pages, 10 figures, 3 tables
References in corpus (9)
- MCMC using Hamiltonian dynamics
- The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data
- Ergodicity of the zigzag process
- The space of ultrametric phylogenetic trees
- Discontinuous Hamiltonian Monte Carlo for discrete parameters and discontinuous likelihoods
- A piecewise deterministic scaling limit of Lifted Metropolis-Hastings in the Curie-Weiss model
- Piecewise Deterministic Markov Processes for Scalable Monte Carlo on Restricted Domains
- Probabilistic Path Hamiltonian Monte Carlo
- Gradients do grow on trees: a linear-time -dimensional gradient for statistical phylogenetics