A Survey of Algorithmic Recourse: Contrastive Explanations and Consequential Recommendations

Machine learning is increasingly used to inform decision making in sensitive situations where decisions have consequential effects on individuals’ lives. In these settings, addition requiring models be accurate and robust, socially relevant values such as fairness, privacy, accountability, explainability play an important role the adoption impact of said technologies. this work, we focus algorithmic recourse , which concerned with providing explanations recommendations individuals who are unfavorably treated by automated decision-making systems. We first perform extensive literature review, align efforts many authors presenting unified definitions, formulations, solutions recourse. Then, provide overview prospective research directions toward community may engage, challenging existing assumptions explicit connections other ethical challenges security, fairness.