Coverage-Validity-Aware Algorithmic Recourse

2023-11-19

arXiv.org

AI Summary

To address the failure of counterfactual recourse under model dynamics, this paper proposes a robust counterfactual explanation method that jointly models coverage and validity. Methodologically, it establishes, for the first time, a unified theoretical framework linking coverage and validity to minimal–maximal probability machines (MPMs), thereby unifying the interpretation of mainstream regularization schemes—including ℓ₂ regularization and class reweighting. Building upon this theory, we design a coverage- and validity-aware linear surrogate model and integrate covariance-robust optimization to generate long-term robust, individualized intervention recommendations for black-box models. Experiments across multiple model drift scenarios demonstrate that our approach improves the long-term validity of counterfactual recommendations by 32.7% on average, while preserving high interpretability and computational efficiency.

Abstract

Algorithmic recourse emerges as a prominent technique to promote the explainability, transparency, and ethics of machine learning models. Existing algorithmic recourse approaches often assume an invariant predictive model; however, the predictive model is usually updated upon the arrival of new data. Thus, a recourse that is valid respective to the present model may become invalid for the future model. To resolve this issue, we propose a novel framework to generate a model-agnostic recourse that exhibits robustness to model shifts. Our framework first builds a coverage-validity-aware linear surrogate of the nonlinear (black-box) model; then, the recourse is generated with respect to the linear surrogate. We establish a theoretical connection between our coverage-validity-aware linear surrogate and the minimax probability machines (MPM). We then prove that by prescribing different covariance robustness, the proposed framework recovers popular regularizations for MPM, including the $ell_2$-regularization and class-reweighting. Furthermore, we show that our surrogate pushes the approximate hyperplane intuitively, facilitating not only robust but also interpretable recourses. The numerical results demonstrate the usefulness and robustness of our framework.

Problem

• Generating robust recourses against model updates
• Ensuring recourse validity under predictive model shifts
• Creating model-agnostic explanations for black-box models

Innovation

• Model-agnostic recourse robust to model shifts
• Coverage-validity-aware linear surrogate for black-box models
• Theoretical connection to minimax probability machines regularization