The YODO algorithm: an efficient computational framework for sensitivity analysis in Bayesian networks

DOI10.1016/J.IJAR.2023.108929MaRDI QIDQ6116525zbMATH OpenOpenAlexFDO

Authors Rafael Ballester-Ripoll, Manuele Leonelli

Publication date 18 July 2023

Published in International Journal of Approximate Reasoning (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2302.00364

zbMATH Keywords

COVID-19sensitivity analysisBayesian networksautomatic differentiation

Mathematics Subject Classification ID

Reasoning under uncertainty in the context of artificial intelligence (68T37)

Abstract: Sensitivity analysis measures the influence of a Bayesian network's parameters on a quantity of interest defined by the network, such as the probability of a variable taking a specific value. Various sensitivity measures have been defined to quantify such influence, most commonly some function of the quantity of interest's partial derivative with respect to the network's conditional probabilities. However, computing these measures in large networks with thousands of parameters can become computationally very expensive. We propose an algorithm combining automatic differentiation and exact inference to efficiently calculate the sensitivity measures in a single pass. It first marginalizes the whole network once, using e.g. variable elimination, and then backpropagates this operation to obtain the gradient with respect to all input parameters. Our method can be used for one-way and multi-way sensitivity analysis and the derivation of admissible regions. Simulation studies highlight the efficiency of our algorithm by scaling it to massive networks with up to 100'000 parameters and investigate the feasibility of generic multi-way analyses. Our routines are also showcased over two medium-sized Bayesian networks: the first modeling the country-risks of a humanitarian crisis, the second studying the relationship between the use of technology and the psychological effects of forced social isolation during the COVID-19 pandemic. An implementation of the methods using the popular machine learning library PyTorch is freely available.

Cites work

• scientific article; zbMATH DE number 867339 (Why is no real title available?)
• A differential approach to inference in Bayesian networks
• A distance measure for bounding probabilistic belief change
• A geometric characterization of sensitivity analysis in monomial models
• Balanced sensitivity functions for tuning multi-dimensional Bayesian network classifiers
• Bayesian network classifiers
• Co-variation for sensitivity analysis in Bayesian networks: properties, consequences and alternatives
• Conditional independence and chain event graphs
• Discrete Bayesian network classifiers: a survey
• Duality of graphical models and tensor networks
• Local Sensitivity of Bayesian Networks to Multiple Simultaneous Parameter Shifts
• Modeling and Reasoning with Bayesian Networks
• Properties of sensitivity analysis of Bayesian belief networks
• Sensitivity analysis beyond linearity
• Sensitivity analysis in multilinear probabilistic models
• Sensitivity analysis of probabilistic networks
• Sensitivity analysis: An aid for belief-network quantification
• Structure-based categorisation of Bayesian network parameters
• Who learns better Bayesian network structures: accuracy and speed of structure learning algorithms

Cited in

(1)

• The diameter of a stochastic matrix: a new measure for sensitivity analysis in Bayesian networks

This page was built for publication: The YODO algorithm: an efficient computational framework for sensitivity analysis in Bayesian networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6116525)