Sampling and estimation on manifolds using the Langevin diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bharath, Karthik, Lewis, Alexander, Sharma, Akash, Tretyakov, Michael V
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917134553055232
author Bharath, Karthik
Lewis, Alexander
Sharma, Akash
Tretyakov, Michael V
author_facet Bharath, Karthik
Lewis, Alexander
Sharma, Akash
Tretyakov, Michael V
contents Error bounds are derived for sampling and estimation using a discretization of an intrinsically defined Langevin diffusion with invariant measure $\text{d}μ_ϕ\propto e^{-ϕ} \mathrm{dvol}_g $ on a compact Riemannian manifold. Two estimators of linear functionals of $μ_ϕ$ based on the discretized Markov process are considered: a time-averaging estimator based on a single trajectory and an ensemble-averaging estimator based on multiple independent trajectories. Imposing no restrictions beyond a nominal level of smoothness on $ϕ$, first-order error bounds, in discretization step size, on the bias and variance/mean-square error of both estimators are derived. The order of error matches the optimal rate in Euclidean and flat spaces, and leads to a first-order bound on distance between the invariant measure $μ_ϕ$ and a stationary measure of the discretized Markov process. This order is preserved even upon using retractions when exponential maps are unavailable in closed form, thus enhancing practicality of the proposed algorithms. Generality of the proof techniques, which exploit links between two partial differential equations and the semigroup of operators corresponding to the Langevin diffusion, renders them amenable for the study of a more general class of sampling algorithms related to the Langevin diffusion. Conditions for extending analysis to the case of non-compact manifolds are discussed. Numerical illustrations with distributions, log-concave and otherwise, on the manifolds of positive and negative curvature elucidate on the derived bounds and demonstrate practical utility of the sampling algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2312_14882
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Sampling and estimation on manifolds using the Langevin diffusion
Bharath, Karthik
Lewis, Alexander
Sharma, Akash
Tretyakov, Michael V
Statistics Theory
Numerical Analysis
Probability
Computation
Machine Learning
Error bounds are derived for sampling and estimation using a discretization of an intrinsically defined Langevin diffusion with invariant measure $\text{d}μ_ϕ\propto e^{-ϕ} \mathrm{dvol}_g $ on a compact Riemannian manifold. Two estimators of linear functionals of $μ_ϕ$ based on the discretized Markov process are considered: a time-averaging estimator based on a single trajectory and an ensemble-averaging estimator based on multiple independent trajectories. Imposing no restrictions beyond a nominal level of smoothness on $ϕ$, first-order error bounds, in discretization step size, on the bias and variance/mean-square error of both estimators are derived. The order of error matches the optimal rate in Euclidean and flat spaces, and leads to a first-order bound on distance between the invariant measure $μ_ϕ$ and a stationary measure of the discretized Markov process. This order is preserved even upon using retractions when exponential maps are unavailable in closed form, thus enhancing practicality of the proposed algorithms. Generality of the proof techniques, which exploit links between two partial differential equations and the semigroup of operators corresponding to the Langevin diffusion, renders them amenable for the study of a more general class of sampling algorithms related to the Langevin diffusion. Conditions for extending analysis to the case of non-compact manifolds are discussed. Numerical illustrations with distributions, log-concave and otherwise, on the manifolds of positive and negative curvature elucidate on the derived bounds and demonstrate practical utility of the sampling algorithm.
title Sampling and estimation on manifolds using the Langevin diffusion
topic Statistics Theory
Numerical Analysis
Probability
Computation
Machine Learning
url https://arxiv.org/abs/2312.14882