Prediction without Preclusion: Recourse Verification with Reachable Sets

TL;DR

This paper introduces recourse verification using reachable sets to identify fixed predictions in machine learning models, highlighting potential preclusion issues in credit and employment decisions.

Contribution

It presents a model-agnostic method for recourse verification with reachable sets, enabling certification of model responsiveness through prediction queries.

Findings

Models can assign fixed predictions, precluding individuals from access.

Reachable sets can certify responsiveness of any model.

Empirical study shows infeasibility of recourse in finance datasets.

Abstract

Machine learning models are often used to decide who receives a loan, a job interview, or a public benefit. Models in such settings use features without considering their actionability. As a result, they can assign predictions that are fixed $-$ meaning that individuals who are denied loans and interviews are, in fact, precluded from access to credit and employment. In this work, we introduce a procedure called recourse verification to test if a model assigns fixed predictions to its decision subjects. We propose a model-agnostic approach for recourse verification with reachable sets $-$ i.e., the set of all points that a person can reach through their actions in feature space. We develop methods to construct reachable sets for discrete feature spaces, which can certify the responsiveness of any model by simply querying its predictions. We conduct a comprehensive empirical study on the infeasibility of recourse on datasets from consumer finance. Our results highlight how models can inadvertently preclude access by assigning fixed predictions and underscore the need to account for actionability in model development.