Variational inference for cutting feedback in misspecified models

DOI10.1214/23-STS886MaRDI QIDQ6181748zbMATH OpenFDO

Authors Xuejun Yu, David J. Nott, Michael Stanley Smith

Publication date 23 January 2024

Published in Statistical Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2108.11066, https://projecteuclid.org/journals/statistical-science/volume-38/issue-3/Variational-Inference-for-Cutting-Feedback-in-Misspecified-Models/10.1214/23-STS886.full

zbMATH Keywords

model misspecificationcutting feedbackBayesian model criticismmodular inference

Mathematics Subject Classification ID

Statistics (62-XX)

Abstract: Bayesian analyses combine information represented by different terms in a joint Bayesian model. When one or more of the terms is misspecified, it can be helpful to restrict the use of information from suspect model components to modify posterior inference. This is called "cutting feedback", and both the specification and computation of the posterior for such "cut models" is challenging. In this paper, we define cut posterior distributions as solutions to constrained optimization problems, and propose optimization-based variational methods for their computation. These methods are faster than existing Markov chain Monte Carlo (MCMC) approaches for computing cut posterior distributions by an order of magnitude. It is also shown that variational methods allow for the evaluation of computationally intensive conflict checks that can be used to decide whether or not feedback should be cut. Our methods are illustrated in a number of simulated and real examples, including an application where recent methodological advances that combine variational inference and MCMC within the variational optimization are used.

Cites work

• scientific article; zbMATH DE number 947414 (Why is no real title available?)
• scientific article; zbMATH DE number 7370574 (Why is no real title available?)
• A General Framework for Updating Belief Distributions
• Adjustment for Missing Confounders Using External Validation Data and Propensity Scores
• Assessment and Adjustment of Approximate Inference Algorithms Using the Law of Total Variance
• Automatic differentiation variational inference
• Bayesian calibration of computer models. (With discussion)
• Checking for prior-data conflict using prior-to-posterior divergences
• Conflict diagnostics in directed acyclic graphs, with applications in Bayesian evidence synthesis
• Cuts in Bayesian graphical models
• Errors-in-Variables in Joint Population Pharmacokinetic/Pharmacodynamic Modeling
• Explaining variational approximations
• Fast and accurate variational inference for models with many latent variables
• Feedback and modularization in a Bayesian meta-analysis of tree traits affecting forest dynamics
• Fixed-form variational posterior approximation through stochastic linear regression
• High-Dimensional Copula Variational Approximation Through Transformation
• Interoperability of statistical models in pandemic preparedness: principles and reality
• Model feedback in Bayesian propensity score estimation
• Modularization in Bayesian analysis, with emphasis on analysis of computer models
• Monte Carlo co-ordinate ascent variational inference
• On the properties of variational approximations of Gibbs posteriors
• Sequential Monte Carlo Samplers
• Stochastic approximation cut algorithm for inference in modularized Bayesian models
• Unbiased Markov Chain Monte Carlo Methods with Couplings
• Variational message passing

This page was built for publication: Variational inference for cutting feedback in misspecified models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6181748)