Nested conformal prediction and quantile out-of-bag ensemble methods

Abstract

Conformal prediction is a popular tool for providing valid prediction sets for classification and regression problems, without relying on any distributional assumptions on the data. While the traditional description of conformal prediction starts with a nonconformity score, we provide an alternate (but equivalent) view that starts with a sequence of nested sets and calibrates them to find a valid prediction set. The nested framework subsumes all nonconformity scores, including recent proposals based on quantile regression and density estimation. While these ideas were originally derived based on sample splitting, our framework seamlessly extends them to other aggregation schemes like cross-conformal, jackknife+ and out-of-bag methods. We use the framework to derive a new algorithm (QOOB, pronounced cube) that combines four ideas: quantile regression, cross-conformalization, ensemble methods and out-of-bag predictions. We develop a computationally efficient implementation of cross-conformal, that is also used by QOOB. In a detailed numerical investigation, QOOB performs either the best or close to the best on all simulated and real datasets. Code for QOOB is available at https://github.com/aigen/QOOB.

38 pages, 4 figures, 8 tables. This version fixes a bug in the proof of Proposition 3. Published paper available at https://www.sciencedirect.com/science/article/abs/pii/S0031320321006725

References in corpus (5)

• A comparison of some conformal quantile regression methods
• Adaptive, Distribution-Free Prediction Intervals for Deep Networks
• Predictive Inference Is Free with the Jackknife+-after-Bootstrap
• Flexible distribution-free conditional predictive bands using density estimators
• Conformal prediction with localization

Cited by in corpus (18)

• Testing for Outliers with Conformal p-values
• Uncertainty Sets for Image Classifiers using Conformal Prediction
• Conformal Prediction using Conditional Histograms
• Selection and Aggregation of Conformal Prediction Sets
• Exchangeability, Conformal Prediction, and Rank Tests
• Distribution-free uncertainty quantification for classification under label shift
• Improving Conditional Coverage via Orthogonal Quantile Regression
• PAC Prediction Sets Under Covariate Shift
• A novel Deep Learning approach for one-step Conformal Prediction approximation
• Calibrated Multiple-Output Quantile Regression with Representation Learning
• Improving Fairness in Criminal Justice Algorithmic Risk Assessments Using Conformal Prediction Sets
• Exact Optimization of Conformal Predictors via Incremental and Decremental Learning
• Sensitivity Analysis of Individual Treatment Effects: A Robust Conformal Inference Approach
• A Comprehensive Framework for Evaluating Time to Event Predictions using the Restricted Mean Survival Time
• Selecting informative conformal prediction sets with false coverage rate control
• Response Quality Assessment for Retrieval-Augmented Generation via Conditional Conformal Factuality
• Selective Ensembles for Consistent Predictions
• AutoCP: Automated Pipelines for Accurate Prediction Intervals