Timezone: »

Bayesian Multi-type Mean Field Multi-agent Imitation Learning
Fan Yang · Alina Vereshchaka · Changyou Chen · Wen Dong

Mon Dec 07 09:00 PM -- 11:00 PM (PST) @ Poster Session 0 #155

Multi-agent Imitation learning (MAIL) refers to the problem that agents learn to perform a task interactively in a multi-agent system through observing and mimicking expert demonstrations, without any knowledge of a reward function from the environment. MAIL has received a lot of attention due to promising results achieved on synthesized tasks, with the potential to be applied to complex real-world multi-agent tasks. Key challenges for MAIL include sample efficiency and scalability. In this paper, we proposed Bayesian multi-type mean field multi-agent imitation learning (BM3IL). Our method improves sample efficiency through establishing a Bayesian formulation for MAIL, and enhances scalability through introducing a new multi-type mean field approximation. We demonstrate the performance of our algorithm through benchmarking with three state-of-the-art multi-agent imitation learning algorithms on several tasks, including solving a multi-agent traffic optimization problem in a real-world transportation network. Experimental results indicate that our algorithm significantly outperforms all other algorithms in all scenarios.

Author Information

Fan Yang (University at Buffalo)
Alina Vereshchaka (University at Buffalo)
Changyou Chen (University at Buffalo)
Wen Dong (University at Buffalo)

Wen Dong is an Assistant Professor of Computer Science and Engineering at the State University of New York at Buffalo with a joint appointment in the Institute of Sustainable Transportation and Logistics. He focuses on modeling human interaction dynamics with stochastic process theory through combining the power of “big data” and the logic/reasoning power of agent-based models, to solve our societies most challenging problems such as transportation sustainability and efficiency. Wen Dong holds a Ph.D. in Media Arts and Sciences from Massachusetts Institute of Technology. His email address is wendong@buffalo.edu.

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors