Aligning Multi-Agent Interactions for Sustainable Food Behavior (AMAI)

Food waste remains a critical environmental challenge, requiring innovative solutions beyond traditional recommendations. This project develops a multi-agent AI system that aligns user behavior with sustainability goals by leveraging reinforcement learning and dynamic adaptation. Unlike static AI suggestions, AMAI introduces interactive agents that collaborate, model user behavior, and adapt interventions based on real-time feedback. By integrating external data sources and delayed rewards, AMAI optimizes long-term adherence to food waste reduction strategies. This research advances AI alignment by shifting from preference-based interactions to goal-driven behavioral alignment.