ACTIONABLE RECOURSE GUIDED BY USER PREFERENCE

Abstract

The growing popularity of machine learning models has led to their increased application in domains directly impacting human lives. In critical fields such as healthcare, banking, and criminal justice, tools that ensure trust and transparency are vital for the responsible adoption of these models. One such tool is actionable recourse (AR) for negatively impacted users. AR describes recommendations of cost-efficient changes to a user's actionable features to help them obtain favorable outcomes. Existing approaches for providing recourse optimize for properties such as proximity, sparsity, validity, and distance-based costs. However, an oftenoverlooked but crucial requirement for actionability is a consideration of User Preference to guide the recourse generation process. Moreover, existing works considering a user's preferences require users to precisely specify their costs for taking actions. This requirement raises questions about the practicality of the corresponding solutions due to the high cognitive loads imposed. In this work, we attempt to capture user preferences via soft constraints in three simple forms: i) scoring continuous features, ii) bounding feature values and iii) ranking categorical features. We propose an optimization framework that is sensitive to user preference and a gradient-based approach to identify User Preferred Actionable Recourse (UP-AR). We empirically demonstrate the proposed approach's superiority in adhering to user preference while maintaining competitive performance in traditional metrics with extensive experiments.

1. INTRODUCTION

Actionable Recourse (AR) (Ustun et al., 2019) is the ability of an individual to obtain the desired outcome from a fixed Machine Learning (ML) model. Several domains such as lending (Siddiqi, 2012) , insurance (Scism, 2019) , resource allocation (Chouldechova et al., 2018; Shroff, 2017) and hiring decisions (Ajunwa et al., 2016) are required to suggest recourses to ensure the trust of the decision system in place; in such scenarios, it is critical to ensure actionability in recourse (otherwise the suggestions are pointless). Consider an individual named Alice who applies for a loan, and the bank, which uses an ML-based classifier, denies it. Naturally, Alice asks -What can I do to get the loan? The inherent question is what action she must take to obtain the loan in the future. Counterfactual explanation introduced in Wachter ( Wachter et al., 2017) provides a what-if scenario to alter the model's decision. AR further aims to provide Alice with a feasible action. A feasible action is both actionable by Alice (meaning she can reasonably execute the directed plan) and suggests as low-cost modifications as possible. While some features (such as age or sex) are inherently inactionable, Alice's personalized constraints may also limit her ability to take action on the suggested recourse (such as a strong reluctance to secure a co-applicant). We call these localized constraints User Preferences, synonymous to userlevel constraints introduced as local feasibility by Mahajan et al. (2019) . Figure 1 illustrates the motivation behind UP-AR. Notice how similar individuals can prefer contrasting recourse. Actionability, as we consider it, is centered explicitly around individual preferences, and similar recourses provided to two individuals (Alice and Bob) with identical feature vectors may not necessarily be equally actionable. Most existing methods of finding actionable recourse are restricted to omission of features from the actionable feature set which Alice does not prefer to act upon, and box constraints (Mothilal et al., 2020) in the form of bounds on feature actions. In this study, we discuss three forms of user preferences and propose a formulation capturing such idiosyncrasies. The proposed score-based preference mechanism can easily communicate with an individual user, providing a seamless feasible recourse. We argue that communicating in terms of preference scores improves the explainability of a recourse generation mechanism, which ultimately improves trust in an ML model. We provide a hypothetical example of UP-AR's ability to adapt to individual preferences in Table 1 . However, there is a lack of contemporary research in understanding Alice's individual preference, which may be contrary to Bob's preference. It is also worth mentioning that the existing method's rigid modifications on actionable features limit the possibility of identifying a recourse due to diminished actionable space. Motivated by the above limitation, we intend to capture soft user preferences along with hard constraints and identify recourse based on local desires without affecting the success rate of identifying recourse. For example, consider Alice prefers to have 80% of fractional cost from loan duration and only 20% from the loan amount, meaning she prefers to have recourse with a minor reduction in the loan amount. Such recourse enables Alice to get the benefits of a loan on her terms. We study the problem of providing user preferred recourse by solving a custom optimization for individual user-based preferences. Our contributions are consolidated as follows: • We start by enabling Alice to provide three types of user preferences: i) Scoring, ii) Ranking, and iii) Bounding. We embed them into an optimization function to guide the recourse generation mechanism. • We then present our approach called User Preferred Actionable Recourse (UP-AR) to identify a recourse instead of overwhelming her with a variety of recourse options. Our approach highlights a cost correction step to address the redundancy induced by our method. • We also consolidate performance metrics with empirical results of UP-AR across multiple datasets and compare them with state-of-art techniques. 1 



Figure 1: Illustration of UP-AR. Similar individuals Alice and Bob with contrasting preferences can have different regions of desired feature space for a recourse.

A hypothetical actionable feature set of adversely affected individuals sharing similar features and corresponding suggested actions by AR and UP-AR. UP-AR provides personalized recourses based on individual user preferences.

AR observes that a feasible recourse achieves good results in all the performance metrics. Existing research focuses on providing feasible recourses, yet comprehensive literature on understanding and incorporating user preferences within the recourse generation mechanism is lacking. It is worth mentioning that instead of better understanding user preferences, Mothilal et al. (2020) provides a user with diverse recourse options and hopes that the user will benefit from at least one. The importance of diverse recourse recommendations have already been explored in recent works(Wachter et al.,

