Saved in:
Bibliographic Details
Main Authors: Boserup, Nicklas, Yang, Gefan, Severinsen, Michael Lind, Hipsley, Christy Anna, Sommer, Stefan
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.08993
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915019344576512
author Boserup, Nicklas
Yang, Gefan
Severinsen, Michael Lind
Hipsley, Christy Anna
Sommer, Stefan
author_facet Boserup, Nicklas
Yang, Gefan
Severinsen, Michael Lind
Hipsley, Christy Anna
Sommer, Stefan
contents We introduce a methodology for performing parameter inference in high-dimensional, non-linear diffusion processes. We illustrate its applicability for obtaining insights into the evolution of and relationships between species, including ancestral state reconstruction. Estimation is performed by utilising score matching to approximate diffusion bridges, which are subsequently used in an importance sampler to estimate log-likelihoods. The entire setup is differentiable, allowing gradient ascent on approximated log-likelihoods. This allows both parameter inference and diffusion mean estimation. This novel, numerically stable, score matching-based parameter inference framework is presented and demonstrated on biological two- and three-dimensional morphometry data.
format Preprint
id arxiv_https___arxiv_org_abs_2411_08993
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Parameter Inference via Differentiable Diffusion Bridge Importance Sampling
Boserup, Nicklas
Yang, Gefan
Severinsen, Michael Lind
Hipsley, Christy Anna
Sommer, Stefan
Machine Learning
We introduce a methodology for performing parameter inference in high-dimensional, non-linear diffusion processes. We illustrate its applicability for obtaining insights into the evolution of and relationships between species, including ancestral state reconstruction. Estimation is performed by utilising score matching to approximate diffusion bridges, which are subsequently used in an importance sampler to estimate log-likelihoods. The entire setup is differentiable, allowing gradient ascent on approximated log-likelihoods. This allows both parameter inference and diffusion mean estimation. This novel, numerically stable, score matching-based parameter inference framework is presented and demonstrated on biological two- and three-dimensional morphometry data.
title Parameter Inference via Differentiable Diffusion Bridge Importance Sampling
topic Machine Learning
url https://arxiv.org/abs/2411.08993