Agent-Oriented Planning in Multi-Agent Systems

2024-10-03

arXiv.org

AI Summary

This work addresses the challenge of efficient task decomposition and collaborative planning for user queries in multi-agent systems. Methodologically, it introduces a meta-agent–centric, agent-oriented planning framework grounded in three formally articulated design principles: solvability, completeness, and non-redundancy. The framework employs a large language model–based meta-agent architecture integrating reward-model–guided task evaluation, lightweight reward modeling, dynamic replanning, and closed-loop feedback. Its key contributions include precise, adaptive subtask decomposition and allocation, as well as runtime performance–driven real-time scheduling optimization. Empirical evaluation on realistic tasks demonstrates a 23.6% improvement in task completion rate and a 31.4% reduction in average response latency compared to both single-agent baselines and state-of-the-art multi-agent approaches.

Technology Category

• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Planning with Language Models
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Through the collaboration of multiple agents possessing diverse expertise and tools, multi-agent systems achieve impressive progress in solving real-world problems. Given the user queries, the meta-agents, serving as the brain within these systems, are required to decompose the queries into multiple sub-tasks that can be allocated to suitable agents capable of solving them, so-called agent-oriented planning. In this study, we identify three critical design principles of agent-oriented planning, including solvability, completeness, and non-redundancy, to ensure that each sub-task is effectively resolved, leading to satisfactory responses to the original queries. These principles further inspire us to propose a novel framework for agent-oriented planning in multi-agent systems, leveraging a fast task decomposition and allocation process followed by an effective and efficient evaluation via a reward model. During the planning process, the meta-agent is also responsible for evaluating the performance of the expert agents, making timely adjustments to the sub-tasks and scheduling as necessary. Besides, we integrate a feedback loop into the proposed framework to further enhance the effectiveness and robustness of such a problem-solving process. Extensive experiments demonstrate the advancement of the proposed framework in solving real-world problems compared to both single-agent systems and existing planning strategies for multi-agent systems.

Problem

• Decompose user queries into solvable sub-tasks
• Ensure sub-task solvability, completeness, and non-redundancy
• Propose AOP framework for efficient task allocation and evaluation

Innovation

• LLM-empowered multi-agent collaboration
• Agent-oriented planning with AOP framework
• Feedback loop enhances problem-solving robustness