Saved in:
| Main Authors: | , , |
|---|---|
| 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 |