Zig-zag sampling for discrete structures and non-reversible phylogenetic MCMC

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Koskela, Jere
Format: Preprint
Published: 2020
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914654733729792
author Koskela, Jere
author_facet Koskela, Jere
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2004_08807
institution arXiv
publishDate 2020
record_format arxiv
spellingShingle Zig-zag sampling for discrete structures and non-reversible phylogenetic MCMC
Koskela, Jere
Computation
Statistics Theory
Populations and Evolution
Methodology
65C05 (Primary) 60J25, 60J95, 62M05, 92D15 (Secondary)
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.
title Zig-zag sampling for discrete structures and non-reversible phylogenetic MCMC
topic Computation
Statistics Theory
Populations and Evolution
Methodology
65C05 (Primary) 60J25, 60J95, 62M05, 92D15 (Secondary)
url https://arxiv.org/abs/2004.08807