Saved in:
Bibliographic Details
Main Authors: Quinzan, Francesco, Casolo, Cecilia, Muandet, Krikamol, Luo, Yucen, Kilbertus, Niki
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2207.09768
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913462436757504
author Quinzan, Francesco
Casolo, Cecilia
Muandet, Krikamol
Luo, Yucen
Kilbertus, Niki
author_facet Quinzan, Francesco
Casolo, Cecilia
Muandet, Krikamol
Luo, Yucen
Kilbertus, Niki
contents Notions of counterfactual invariance (CI) have proven essential for predictors that are fair, robust, and generalizable in the real world. We propose graphical criteria that yield a sufficient condition for a predictor to be counterfactually invariant in terms of a conditional independence in the observational distribution. In order to learn such predictors, we propose a model-agnostic framework, called Counterfactually Invariant Prediction (CIP), building on the Hilbert-Schmidt Conditional Independence Criterion (HSCIC), a kernel-based conditional dependence measure. Our experimental results demonstrate the effectiveness of CIP in enforcing counterfactual invariance across various simulated and real-world datasets including scalar and multi-variate settings.
format Preprint
id arxiv_https___arxiv_org_abs_2207_09768
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Learning Counterfactually Invariant Predictors
Quinzan, Francesco
Casolo, Cecilia
Muandet, Krikamol
Luo, Yucen
Kilbertus, Niki
Machine Learning
Notions of counterfactual invariance (CI) have proven essential for predictors that are fair, robust, and generalizable in the real world. We propose graphical criteria that yield a sufficient condition for a predictor to be counterfactually invariant in terms of a conditional independence in the observational distribution. In order to learn such predictors, we propose a model-agnostic framework, called Counterfactually Invariant Prediction (CIP), building on the Hilbert-Schmidt Conditional Independence Criterion (HSCIC), a kernel-based conditional dependence measure. Our experimental results demonstrate the effectiveness of CIP in enforcing counterfactual invariance across various simulated and real-world datasets including scalar and multi-variate settings.
title Learning Counterfactually Invariant Predictors
topic Machine Learning
url https://arxiv.org/abs/2207.09768