Generalized conformal predictive systems are extended to non-exchangeable settings under distributional shifts via permutation weights and robust weight-uncertainty boxes with finite-sample or asymptotic guarantees.
Venn-abers predictors.arXiv preprint arXiv:1211.0025
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
This paper continues study, both theoretical and empirical, of the method of Venn prediction, concentrating on binary prediction problems. Venn predictors produce probability-type predictions for the labels of test objects which are guaranteed to be well calibrated under the standard assumption that the observations are generated independently from the same distribution. We give a simple formalization and proof of this property. We also introduce Venn-Abers predictors, a new class of Venn predictors based on the idea of isotonic regression, and report promising empirical results both for Venn-Abers predictors and for their more computationally efficient simplified version.
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 4roles
background 1polarities
background 1representative citing papers
CASCADE propagates Venn-Abers epistemic uncertainty from a screening classifier to adapt non-conformity scores and produce narrower conformal intervals for high-confidence regression cases in PD medication forecasting.
Post-hoc calibration of miscalibrated black-box predictions on a labeled sample improves efficiency of prediction-powered inference for semisupervised mean estimation.
Pith review generated a malformed one-line summary.
citing papers explorer
-
Generalized Conformal Predictive Systems Under Distributional Shifts
Generalized conformal predictive systems are extended to non-exchangeable settings under distributional shifts via permutation weights and robust weight-uncertainty boxes with finite-sample or asymptotic guarantees.
-
Calibeating Prediction-Powered Inference
Post-hoc calibration of miscalibrated black-box predictions on a labeled sample improves efficiency of prediction-powered inference for semisupervised mean estimation.