Amortized Generation of Sequential Algorithmic Recourses for Black-box Models

TL;DR

This paper introduces a novel stochastic-control-based method for generating sequential algorithmic recourses for black-box models, enabling step-by-step actionable feedback that is model-agnostic, amortized, and respects data and causal constraints.

Contribution

It proposes a new approach for sequential ARs that is amortized, model-agnostic, and capable of incorporating data manifold, causal, and sparsity constraints.

Findings

Successfully generates sequential ARs on real-world datasets.

Recourses respect data manifold and causal relations.

Method is applicable to multiple data points without re-optimization.

Abstract

Explainable machine learning (ML) has gained traction in recent years due to the increasing adoption of ML-based systems in many sectors. Algorithmic Recourses (ARs) provide "what if" feedback of the form "if an input datapoint were x' instead of x, then an ML-based system's output would be y' instead of y." ARs are attractive due to their actionable feedback, amenability to existing legal frameworks, and fidelity to the underlying ML model. Yet, current AR approaches are single shot -- that is, they assume x can change to x' in a single time period. We propose a novel stochastic-control-based approach that generates sequential ARs, that is, ARs that allow x to move stochastically and sequentially across intermediate states to a final state x'. Our approach is model agnostic and black box. Furthermore, the calculation of ARs is amortized such that once trained, it applies to multiple datapoints without the need for re-optimization. In addition to these primary characteristics, our approach admits optional desiderata such as adherence to the data manifold, respect for causal relations, and sparsity -- identified by past research as desirable properties of ARs. We evaluate our approach using three real-world datasets and show successful generation of sequential ARs that respect other recourse desiderata.