Saved in:
Bibliographic Details
Main Authors: Ballester-Ripoll, Rafael, Leonelli, Manuele
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.05764
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914830114357248
author Ballester-Ripoll, Rafael
Leonelli, Manuele
author_facet Ballester-Ripoll, Rafael
Leonelli, Manuele
contents Traditionally, the sensitivity analysis of a Bayesian network studies the impact of individually modifying the entries of its conditional probability tables in a one-at-a-time (OAT) fashion. However, this approach fails to give a comprehensive account of each inputs' relevance, since simultaneous perturbations in two or more parameters often entail higher-order effects that cannot be captured by an OAT analysis. We propose to conduct global variance-based sensitivity analysis instead, whereby $n$ parameters are viewed as uncertain at once and their importance is assessed jointly. Our method works by encoding the uncertainties as $n$ additional variables of the network. To prevent the curse of dimensionality while adding these dimensions, we use low-rank tensor decomposition to break down the new potentials into smaller factors. Last, we apply the method of Sobol to the resulting network to obtain $n$ global sensitivity indices. Using a benchmark array of both expert-elicited and learned Bayesian networks, we demonstrate that the Sobol indices can significantly differ from the OAT indices, thus revealing the true influence of uncertain parameters and their interactions.
format Preprint
id arxiv_https___arxiv_org_abs_2406_05764
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Global Sensitivity Analysis of Uncertain Parameters in Bayesian Networks
Ballester-Ripoll, Rafael
Leonelli, Manuele
Artificial Intelligence
Traditionally, the sensitivity analysis of a Bayesian network studies the impact of individually modifying the entries of its conditional probability tables in a one-at-a-time (OAT) fashion. However, this approach fails to give a comprehensive account of each inputs' relevance, since simultaneous perturbations in two or more parameters often entail higher-order effects that cannot be captured by an OAT analysis. We propose to conduct global variance-based sensitivity analysis instead, whereby $n$ parameters are viewed as uncertain at once and their importance is assessed jointly. Our method works by encoding the uncertainties as $n$ additional variables of the network. To prevent the curse of dimensionality while adding these dimensions, we use low-rank tensor decomposition to break down the new potentials into smaller factors. Last, we apply the method of Sobol to the resulting network to obtain $n$ global sensitivity indices. Using a benchmark array of both expert-elicited and learned Bayesian networks, we demonstrate that the Sobol indices can significantly differ from the OAT indices, thus revealing the true influence of uncertain parameters and their interactions.
title Global Sensitivity Analysis of Uncertain Parameters in Bayesian Networks
topic Artificial Intelligence
url https://arxiv.org/abs/2406.05764