The research direction of multi-agent systems is an interdisciplinary field that combines deep reinforcement learning, multi-agent systems, and game theory. It primarily addresses the problem of how each intelligent agent in a complex environment with multiple decision-making entities can optimize its utility function and find optimal strategies through cooperation, competition, and mixed game scenarios through interactive gameplay between agents. It is an important approach to exploring collective intelligence and a pathway towards achieving the ultimate goal of the next generation of artificial intelligence.

The main research areas include: general single-agent and multi-agent reinforcement learning theory, methods, and applications in open environments; intelligent group robotics and engineering applications; design, analysis, and algorithmic research on game environments and game agents; intelligent agent modeling and artificial intelligence algorithms in computational economics; artificial intelligence decision-making methods for supply chains; reinforcement learning methods based on risk measurement incorporating behavioral characteristics; modeling, analysis, simulation, prediction, optimization, and control of large-scale, intelligent, networked, multi-level, multi-scale, nonlinear, uncertain, and time-varying dynamic systems.

The research has been supported by funding from national, provincial, international cooperation, and industry projects, including national key R&D projects, the Ministry of Science and Technology's AI 2030 key R&D program, and major and key projects of the National Natural Science Foundation. The related achievements have been published in top conferences and journals in the field of machine learning, with hundreds of papers published and dozens of patents filed. The team has received recognition for their contributions in multi-agent systems, including the Best System Paper Award at the Conference on Robot Learning (CoRL) 2020, the Most Visionary Paper Award at the International Conference on Autonomous Agents and Multiagent Systems (AAMAS) 2021, the WAIC Cloud Sail Award (one of only ten recipients in 2022) at the World Artificial Intelligence Conference (WAIC), the ACM SIGAI China Rising Star Award, and the Huawei UK Best Technical Breakthrough Award.