Meta-Uncertainty in Bayesian Model Comparison
Schmitt, Marvin, Radev, Stefan T., Bürkner, Paul-Christian
–arXiv.org Artificial Intelligence
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.
arXiv.org Artificial Intelligence
Feb-21-2023
- Country:
- Europe
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Spain > Valencian Community
- Valencia Province > Valencia (0.04)
- Germany > Baden-Württemberg
- Stuttgart Region > Stuttgart (0.04)
- United Kingdom > England
- Europe
- Genre:
- Research Report (1.00)
- Industry:
- Health & Medicine > Therapeutic Area > Immunology (0.93)