TPMARL-MSS Thought-Perception Multi-Agent Reinforcement Learning Meeting Scheduling System

In today's evolving society, the increasing complexity and frequency of meetings necessitate advanced scheduling systems. Traditional methods are constrained by rigid prede-fined strategies, lack intelligent negotiation mechanisms, and often compromise user privacy. Addressing these challenges, we introduce the Thought-Perception Multi-Agent Reinforcement Learning Meeting Scheduling System (TPMARL-MSS). Unlike conventional systems, TPMARL-MSS autonomously learns and refines its strategies through continuous feedback. It features automated negotiation and adaptive decision-making, offering a more nuanced scheduling approach. Importantly, our Thought-Perception module protect privacy, allowing the agent to deduce preferences from participant behavior without revealing personal data. Evaluations on the real-world dataset shows that TPMARL-MSS surpasses traditional methods in efficiency and schedule quality, highlighting its practical applicability.