aiqbench analysis
This paper extends Venn-Abers probabilistic predictors from binary classification and bounded regression to unbounded regression by integrating conformal prediction methods. Empirical results suggest these generalized regressors offer improved predictive efficiency over standard models when training datasets are sufficiently large. The work advances theoretical machine learning frameworks for uncertainty quantification in continuous output spaces.
Inductive Venn-Abers and related regressors
arXiv:2605.06646v2 Announce Type: replace Abstract: Venn-Abers predictors are probabilistic predictors that enjoy appealing properties of validity, but their major limitation is that they have been applicable only to binary classification, apart from a recent extension to bounded regression. We generalize them to the case of unbounded regression, which requires adding an element of conformal prediction. In our simulation and empirical studies we investigate the predictive efficiency of point regressors derived from Venn-Abers regressors and argue that they somewhat improve the predictive efficiency of standard regressors for larger training sets.
Abstract:Venn-Abers predictors are probabilistic predictors that enjoy appealing properties of validity, but their major limitation is that they have been applicable only to binary classification, apart from a recent extension to bounded regression. We generalize them to the case of unbounded regression, which requires adding an element of conformal prediction. In our simulation and empirical studies we investigate the predictive efficiency of point regressors derived from Venn-Abers regressors and argue that they somewhat improve the predictive efficiency of standard regressors for larger training sets.
Machine Learning (cs.LG)
arXiv:2605.06646 [cs.LG]
(or arXiv:2605.06646v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2605.06646 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 176 words
Related coverage
SPACR: Single-Pass Adaptive Training of Uncertainty-Aware Conformal Regressors
arXiv · Sep 4, 2026
SpikeSSL: A Universal Spike Inference Framework with Dynamics-Informed State-Space Layers
arXiv · Oct 10, 2026
The Next Swan: Frank Ramsey, Variable Hypotheticals, and the Bet on Induction
LessWrong · Jun 8, 2026
QORH: Learning Query-Specific Ordered Rule Hypergraphs for Inductive Knowledge Graph Link Prediction
arXiv · Oct 15, 2026