Conformal Classification with Equalized Coverage for Adaptively Selected Groups

2024/05/23 by Yanfei Zhou, Matteo Sesia, Zhou, Yanfei +1 · 6 citations
Mathematics · #Numerical methods in inverse problems #Advanced Research in Science and Engineering #Statistical Methods and Inference

paper · pdf · doi:10.48550/arxiv.2405.15106

Abstract

This paper introduces a conformal inference method to evaluate uncertainty in classification by generating prediction sets with valid coverage conditional on adaptively chosen features. These features are carefully selected to reflect potential model limitations or biases. This can be useful to find a practical compromise between efficiency -- by providing informative predictions -- and algorithmic fairness -- by ensuring equalized coverage for the most sensitive groups. We demonstrate the validity and effectiveness of this method on simulated and real data sets.

Cited by

• Interpretable Multivariate Conformal Prediction with Fast Transductive Standardization
• Set to Be Fair: Demographic Parity Constraints for Set-Valued Classification
• Calibrated Counterfactual Conformal Fairness (C³F): Post-hoc, Shift-Aware Coverage Parity via Conformal Prediction and Counterfactual Regularization
• FedCF: Fair Federated Conformal Prediction
• Smoothing-Based Conformal Prediction for Balancing Efficiency and Interpretability
• Conformal Prediction Meets Long-tail Classification

Related

• Noise-Adaptive Conformal Classification with Marginal Coverage
• Adaptive Coverage Policies in Conformal Prediction
• Conformal Prediction Sets with Improved Conditional Coverage using Trust Scores
• Adaptive Uncertainty Quantification for Generative AI
• Conformal Prediction With Conditional Guarantees