Towards User Guided Actionable Recourse

Abstract

Machine Learning's proliferation in critical fields such as healthcare, banking, and criminal justice has motivated the creation of tools which ensure trust and transparency in ML 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 often-overlooked but crucial requirement for actionability is a consideration of User Preference to guide the recourse generation process. 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. Finally, we propose a gradient-based approach to identify User Preferred Actionable Recourse (UP-AR). We carried out extensive experiments to verify the effectiveness of our approach.

References in corpus (13)

• Explaining and Harnessing Adversarial Examples
• Intriguing properties of neural networks
• On Calibration of Modern Neural Networks
• Explaining Machine Learning Classifiers through Diverse Counterfactual Explanations
• Actionable Recourse in Linear Classification
• FACE: Feasible and Actionable Counterfactual Explanations
• Learning Model-Agnostic Counterfactual Explanations for Tabular Data
• Preserving Causal Constraints in Counterfactual Explanations for Machine Learning Classifiers
• Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
• Getting a CLUE: A Method for Explaining Uncertainty Estimates
• Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses
• Issues with post-hoc counterfactual explanations: a discussion
• Personalized Algorithmic Recourse with Preference Elicitation