Algorithmic Recourse in Partially and Fully Confounded Settings Through Bounding Counterfactual Effects

TL;DR

This paper introduces a novel causal approach for algorithmic recourse that accounts for unobserved confounding and relaxes traditional assumptions, providing bounds on counterfactual effects to ensure actionable recourse.

Contribution

It extends causal recourse methods to settings with hidden confounding, using only causal graph and confounding structure to bound effects without strict modeling assumptions.

Findings

Provides bounds on counterfactual effects under confounding.

Guarantees recourse when lower bounds exceed decision thresholds.

Relaxes assumptions of no hidden confounding in causal models.

Abstract

Algorithmic recourse aims to provide actionable recommendations to individuals to obtain a more favourable outcome from an automated decision-making system. As it involves reasoning about interventions performed in the physical world, recourse is fundamentally a causal problem. Existing methods compute the effect of recourse actions using a causal model learnt from data under the assumption of no hidden confounding and modelling assumptions such as additive noise. Building on the seminal work of Balke and Pearl (1994), we propose an alternative approach for discrete random variables which relaxes these assumptions and allows for unobserved confounding and arbitrary structural equations. The proposed approach only requires specification of the causal graph and confounding structure and bounds the expected counterfactual effect of recourse actions. If the lower bound is above a certain threshold, i.e., on the other side of the decision boundary, recourse is guaranteed in expectation.