Bayesian updating is reframed as an optimization problem without inherent uncertainty quantification, and a PAC-style calibration step is proposed to give predictive intervals frequentist coverage.
Safe learning: bridging the gap between bayes, mdl and statistical learning theory via empirical convexity
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Position: There Is No Free Bayesian Uncertainty Quantification
Bayesian updating is reframed as an optimization problem without inherent uncertainty quantification, and a PAC-style calibration step is proposed to give predictive intervals frequentist coverage.