Calibrated Counterfactual Conformal Fairness ($C^3F$): Post-hoc, Shift-Aware Coverage Parity via Conformal Prediction and Counterfactual Regularization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Alpay, Faruk, Alpay, Taylan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909815093067776
author Alpay, Faruk
Alpay, Taylan
author_facet Alpay, Faruk
Alpay, Taylan
contents We present Calibrated Counterfactual Conformal Fairness ($C^3F$), a post-hoc procedure that targets group-conditional coverage parity under covariate shift. $C^3F$ combines importance-weighted conformal calibration with a counterfactual regularizer based on path-specific effects in a structural causal model. The method estimates group-specific nonconformity quantiles using likelihood-ratio weights so that coverage degrades gracefully with the second moment of the weights. We derive finite-sample lower bounds on group-wise coverage and a bound on the equalized conditional coverage gap, and we show first-order control of a counterfactual coverage-parity surrogate via smooth threshold regularization. The approach is model-agnostic, label-efficient, and deployable without retraining. Empirical evaluations on standard classification benchmarks demonstrate improved group-conditional coverage and competitive efficiency relative to shift-aware and fairness-oriented conformal baselines. We discuss practical considerations, including partial availability of sensitive attributes and robustness to structural causal misspecification.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25295
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Calibrated Counterfactual Conformal Fairness ($C^3F$): Post-hoc, Shift-Aware Coverage Parity via Conformal Prediction and Counterfactual Regularization
Alpay, Faruk
Alpay, Taylan
Methodology
62G15, 62G08, 62C12, 62H30, 62G20
G.3; I.5.1; I.5.2
We present Calibrated Counterfactual Conformal Fairness ($C^3F$), a post-hoc procedure that targets group-conditional coverage parity under covariate shift. $C^3F$ combines importance-weighted conformal calibration with a counterfactual regularizer based on path-specific effects in a structural causal model. The method estimates group-specific nonconformity quantiles using likelihood-ratio weights so that coverage degrades gracefully with the second moment of the weights. We derive finite-sample lower bounds on group-wise coverage and a bound on the equalized conditional coverage gap, and we show first-order control of a counterfactual coverage-parity surrogate via smooth threshold regularization. The approach is model-agnostic, label-efficient, and deployable without retraining. Empirical evaluations on standard classification benchmarks demonstrate improved group-conditional coverage and competitive efficiency relative to shift-aware and fairness-oriented conformal baselines. We discuss practical considerations, including partial availability of sensitive attributes and robustness to structural causal misspecification.
title Calibrated Counterfactual Conformal Fairness ($C^3F$): Post-hoc, Shift-Aware Coverage Parity via Conformal Prediction and Counterfactual Regularization
topic Methodology
62G15, 62G08, 62C12, 62H30, 62G20
G.3; I.5.1; I.5.2
url https://arxiv.org/abs/2509.25295