English

Meta-Uncertainty in Bayesian Model Comparison

Machine Learning 2023-02-22 v3 Machine Learning

Abstract

Bayesian model comparison (BMC) offers a principled probabilistic approach to study and rank competing models. In standard BMC, we construct a discrete probability distribution over the set of possible models, conditional on the observed data of interest. These posterior model probabilities (PMPs) are measures of uncertainty, but -- when derived from a finite number of observations -- are also uncertain themselves. In this paper, we conceptualize distinct levels of uncertainty which arise in BMC. We explore a fully probabilistic framework for quantifying meta-uncertainty, resulting in an applied method to enhance any BMC workflow. Drawing on both Bayesian and frequentist techniques, we represent the uncertainty over the uncertain PMPs via meta-models which combine simulated and observed data into a predictive distribution for PMPs on new data. We demonstrate the utility of the proposed method in the context of conjugate Bayesian regression, likelihood-based inference with Markov chain Monte Carlo, and simulation-based inference with neural networks.

Keywords

Cite

@article{arxiv.2210.07278,
  title  = {Meta-Uncertainty in Bayesian Model Comparison},
  author = {Marvin Schmitt and Stefan T. Radev and Paul-Christian Bürkner},
  journal= {arXiv preprint arXiv:2210.07278},
  year   = {2023}
}

Comments

accepted at AISTATS 2023

R2 v1 2026-06-28T03:35:13.487Z