Prediction-Centric Uncertainty Quantification via MMD

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Zheyang, Knoblauch, Jeremias, Power, Sam, Oates, Chris. J.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915207117275136
author Shen, Zheyang
Knoblauch, Jeremias
Power, Sam
Oates, Chris. J.
author_facet Shen, Zheyang
Knoblauch, Jeremias
Power, Sam
Oates, Chris. J.
contents Deterministic mathematical models, such as those specified via differential equations, are a powerful tool to communicate scientific insight. However, such models are necessarily simplified descriptions of the real world. Generalised Bayesian methodologies have been proposed for inference with misspecified models, but these are typically associated with vanishing parameter uncertainty as more data are observed. In the context of a misspecified deterministic mathematical model, this has the undesirable consequence that posterior predictions become deterministic and certain, while being incorrect. Taking this observation as a starting point, we propose Prediction-Centric Uncertainty Quantification, where a mixture distribution based on the deterministic model confers improved uncertainty quantification in the predictive context. Computation of the mixing distribution is cast as a (regularised) gradient flow of the maximum mean discrepancy (MMD), enabling consistent numerical approximations to be obtained. Results are reported on both a toy model from population ecology and a real model of protein signalling in cell biology.
format Preprint
id arxiv_https___arxiv_org_abs_2410_11637
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Prediction-Centric Uncertainty Quantification via MMD
Shen, Zheyang
Knoblauch, Jeremias
Power, Sam
Oates, Chris. J.
Methodology
Deterministic mathematical models, such as those specified via differential equations, are a powerful tool to communicate scientific insight. However, such models are necessarily simplified descriptions of the real world. Generalised Bayesian methodologies have been proposed for inference with misspecified models, but these are typically associated with vanishing parameter uncertainty as more data are observed. In the context of a misspecified deterministic mathematical model, this has the undesirable consequence that posterior predictions become deterministic and certain, while being incorrect. Taking this observation as a starting point, we propose Prediction-Centric Uncertainty Quantification, where a mixture distribution based on the deterministic model confers improved uncertainty quantification in the predictive context. Computation of the mixing distribution is cast as a (regularised) gradient flow of the maximum mean discrepancy (MMD), enabling consistent numerical approximations to be obtained. Results are reported on both a toy model from population ecology and a real model of protein signalling in cell biology.
title Prediction-Centric Uncertainty Quantification via MMD
topic Methodology
url https://arxiv.org/abs/2410.11637