A0783
Title: Learning to make adherence-aware recommendations
Authors: Guanting Chen - University of North Carolina at Chapel Hill (United States) [presenting]
Abstract: As AI systems continue to make recommendations for human decision-making, it is frequently observed that human agents sometimes disregard these recommendations. In such cases, it may be beneficial for the AI system to refrain from providing the "optimal" recommendation, which assumes perfect adherence from the agent. A proposed decision-making model considers adherence-aware recommendations, accounting for the varying levels of adherence exhibited by human agents across different states and actions. Aside from the model, accountable and near-optimal reinforcement learning algorithms specifically designed to address adherence-aware recommendations are also introduced.
