Learning Counterfactually Invariant Predictors

2022/07/20 by Quinzan, Francesco, Casolo, Cecilia, Muandet, Krikamol +2 · 2 citations
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2207.09768

Abstract

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.

Cited by

• DISCO: Mitigating Bias in Deep Learning with Conditional Distance Correlation
• When Shift Happens - Confounding Is to Blame

Related

• Results on Counterfactual Invariance
• Double Robust Representation Learning for Counterfactual Prediction
• From Invariant Representations to Invariant Data: Provable Robustness to Spurious Correlations via Noisy Counterfactual Matching
• Exogenous Matching: Learning Good Proposals for Tractable Counterfactual Estimation
• Synthetic Counterfactual Labels for Efficient Conformal Counterfactual Inference