Empowering Bayes with sources of uncertainty: from confidence modelling to sparse estimation
Abstract
We introduce sources of uncertainty (SoU) inference, a general framework that extends Bayesian inference by allowing researchers to explicitly encode heterogeneous confidence across model components. This heterogeneity allows the probabilistic treatment to vary selectively across the model, controlling both the uncertainty retained within each component and its propagation to the rest of the model. Empirical Bayes and profiling-type inference emerge as limiting configurations of this selective treatment. Because confidence acts through a mechanism distinct from prior informativeness, the two can be controlled separately, providing an additional degree of freedom for Bayesian model design. Building on this selective treatment, we develop a general approach for inducing sparsity in statistical models and illustrate its use in linear and logistic regression, as well as in Bayesian neural networks, where it enables self-pruning.
Related
- Performance uncertainty in medical image analysis: a large-scale investigation of confidence intervals
- Position--momentum uncertainty relation for the average confidence width
- Adaptive Confidence Sets for Nonparametric Density Estimation under Local Differential Privacy
- Advance versus Spot Selling in Queueing Systems: the Effect of Uncertainty Sources
- Selection instability and coefficient uncertainty in Cox modelling after Lasso selection
BibTeX
@misc{rosa2026empowering,
title = {{Empowering Bayes with sources of uncertainty: from confidence modelling to sparse estimation}},
author = {Rafael Mouallem Rosa and Julyan Arbel and Hien Duy Nguyen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2605.03134},
eprint = {2605.03134},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.03134},
}
RIS
TY - UNPB
TI - Empowering Bayes with sources of uncertainty: from confidence modelling to sparse estimation
AU - Rafael Mouallem Rosa
AU - Julyan Arbel
AU - Hien Duy Nguyen
PY - 2026
DA - 2026/10/16
PB - arXiv
DO - 10.48550/arxiv.2605.03134
UR - https://arxiv.org/abs/2605.03134
AB - We introduce sources of uncertainty (SoU) inference, a general framework that extends Bayesian inference by allowing researchers to explicitly encode heterogeneous confidence across model components. This heterogeneity allows the probabilistic treatment to vary selectively across the model, controlling both the uncertainty retained within each component and its propagation to the rest of the model. Empirical Bayes and profiling-type inference emerge as limiting configurations of this selective treatment. Because confidence acts through a mechanism distinct from prior informativeness, the two can be controlled separately, providing an additional degree of freedom for Bayesian model design. Building on this selective treatment, we develop a general approach for inducing sparsity in statistical models and illustrate its use in linear and logistic regression, as well as in Bayesian neural networks, where it enables self-pruning.
ER -