Decomposing Counterfactual Explanations for Consequential Decision Making

TL;DR

This paper introduces DEAR, a novel framework for algorithmic recourse that effectively balances the assumptions of feature independence and causal dependency, enabling reliable and low-cost decision reversal in complex domains.

Contribution

DEAR disentangles co-varying features from promising recourse features, bridging the gap between independent manipulability and causal models for practical recourse generation.

Findings

DEAR provides reliable, low-cost recourse in real-world data.

It effectively handles feature dependencies without strong causal assumptions.

Experiments validate the theoretical advantages of the proposed framework.

Abstract

The goal of algorithmic recourse is to reverse unfavorable decisions (e.g., from loan denial to approval) under automated decision making by suggesting actionable feature changes (e.g., reduce the number of credit cards). To generate low-cost recourse the majority of methods work under the assumption that the features are independently manipulable (IMF). To address the feature dependency issue the recourse problem is usually studied through the causal recourse paradigm. However, it is well known that strong assumptions, as encoded in causal models and structural equations, hinder the applicability of these methods in complex domains where causal dependency structures are ambiguous. In this work, we develop \texttt{DEAR} (DisEntangling Algorithmic Recourse), a novel and practical recourse framework that bridges the gap between the IMF and the strong causal assumptions. \texttt{DEAR} generates recourses by disentangling the latent representation of co-varying features from a subset of promising recourse features to capture the main practical recourse desiderata. Our experiments on real-world data corroborate our theoretically motivated recourse model and highlight our framework's ability to provide reliable, low-cost recourse in the presence of feature dependencies.