Saved in:
Bibliographic Details
Main Authors: Leviyev, Alex, Iacovelli, Francesco, Zimmerman, Aaron
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.01942
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914017189036032
author Leviyev, Alex
Iacovelli, Francesco
Zimmerman, Aaron
author_facet Leviyev, Alex
Iacovelli, Francesco
Zimmerman, Aaron
contents Bayesian inference plays a central role in scientific and engineering applications by enabling principled reasoning under uncertainty. However, sampling from generic probability distributions remains a computationally demanding task. This difficulty is compounded when the distributions are ill-conditioned, multi-modal, or supported on topologically non-Euclidean spaces. Motivated by challenges in gravitational wave parameter estimation, we propose simulating a Langevin diffusion augmented with a birth-death process. The dynamics are rescaled with a simple preconditioner, and generalized to apply to the product spaces of a hypercube and hypertorus. Our method is first-order and embarrassingly parallel with respect to model evaluations, making it well-suited for algorithmic differentiation and modern hardware accelerators. We validate the algorithm on a suite of toy problems and successfully apply it to recover the parameters of GW150914 -- the first observed binary black hole merger. This approach addresses key limitations of traditional sampling methods, and introduces a template that can be used to design robust samplers in the future.
format Preprint
id arxiv_https___arxiv_org_abs_2509_01942
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Bayesian Sampling with Langevin Birth-Death Dynamics
Leviyev, Alex
Iacovelli, Francesco
Zimmerman, Aaron
Applications
General Relativity and Quantum Cosmology
62F15, 65C05, 85-04, 62H11
G.3
Bayesian inference plays a central role in scientific and engineering applications by enabling principled reasoning under uncertainty. However, sampling from generic probability distributions remains a computationally demanding task. This difficulty is compounded when the distributions are ill-conditioned, multi-modal, or supported on topologically non-Euclidean spaces. Motivated by challenges in gravitational wave parameter estimation, we propose simulating a Langevin diffusion augmented with a birth-death process. The dynamics are rescaled with a simple preconditioner, and generalized to apply to the product spaces of a hypercube and hypertorus. Our method is first-order and embarrassingly parallel with respect to model evaluations, making it well-suited for algorithmic differentiation and modern hardware accelerators. We validate the algorithm on a suite of toy problems and successfully apply it to recover the parameters of GW150914 -- the first observed binary black hole merger. This approach addresses key limitations of traditional sampling methods, and introduces a template that can be used to design robust samplers in the future.
title Efficient Bayesian Sampling with Langevin Birth-Death Dynamics
topic Applications
General Relativity and Quantum Cosmology
62F15, 65C05, 85-04, 62H11
G.3
url https://arxiv.org/abs/2509.01942