With the rapid deployment of algorithms in high-stakes domains such as finance, healthcare, and justice, recourse has emerged as a critical tool in explainable AI. Algorithmic Recourse provides individuals with recommendations to reverse unfavorable outcomes, addressing the growing need for transparency and accountability in automated decision-making systems. My research investigates the design of recourse methods that are not only effective but also fair, robust, and grounded in ethical considerations.
Kshitij Kayastha (Wed,) studied this question.