Saved in:
Bibliographic Details
Main Authors: Rosato, Conor, Murphy, Joshua, Varsi, Alessandro, Horridge, Paul, Maskell, Simon
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.17296
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910540320735232
author Rosato, Conor
Murphy, Joshua
Varsi, Alessandro
Horridge, Paul
Maskell, Simon
author_facet Rosato, Conor
Murphy, Joshua
Varsi, Alessandro
Horridge, Paul
Maskell, Simon
contents Sequential Monte Carlo Squared (SMC$^2$) is a Bayesian method which can infer the states and parameters of non-linear, non-Gaussian state-space models. The standard random-walk proposal in SMC$^2$ faces challenges, particularly with high-dimensional parameter spaces. This study outlines a novel approach by harnessing first-order gradients derived from a Common Random Numbers - Particle Filter (CRN-PF) using PyTorch. The resulting gradients can be leveraged within a Langevin proposal without accept/reject. Including Langevin dynamics within the proposal can result in a higher effective sample size and more accurate parameter estimates when compared with the random-walk. The resulting algorithm is parallelized on distributed memory using Message Passing Interface (MPI) and runs in $\mathcal{O}(\log_2N)$ time complexity. Utilizing 64 computational cores we obtain a 51x speed-up when compared to a single core. A GitHub link is given which provides access to the code.
format Preprint
id arxiv_https___arxiv_org_abs_2407_17296
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhanced SMC$^2$: Leveraging Gradient Information from Differentiable Particle Filters Within Langevin Proposals
Rosato, Conor
Murphy, Joshua
Varsi, Alessandro
Horridge, Paul
Maskell, Simon
Machine Learning
Applications
Sequential Monte Carlo Squared (SMC$^2$) is a Bayesian method which can infer the states and parameters of non-linear, non-Gaussian state-space models. The standard random-walk proposal in SMC$^2$ faces challenges, particularly with high-dimensional parameter spaces. This study outlines a novel approach by harnessing first-order gradients derived from a Common Random Numbers - Particle Filter (CRN-PF) using PyTorch. The resulting gradients can be leveraged within a Langevin proposal without accept/reject. Including Langevin dynamics within the proposal can result in a higher effective sample size and more accurate parameter estimates when compared with the random-walk. The resulting algorithm is parallelized on distributed memory using Message Passing Interface (MPI) and runs in $\mathcal{O}(\log_2N)$ time complexity. Utilizing 64 computational cores we obtain a 51x speed-up when compared to a single core. A GitHub link is given which provides access to the code.
title Enhanced SMC$^2$: Leveraging Gradient Information from Differentiable Particle Filters Within Langevin Proposals
topic Machine Learning
Applications
url https://arxiv.org/abs/2407.17296