Attend LIONS Seminar with guest Chicheng Zhang, March 6

About:

Chicheng Zhang is an assistant professor in the Department of Computer Science at the University of Arizona. He is interested in theory and applications of interactive machine learning, such as active learning, reinforcement learning and imitation learning. He regularly serves as a program committee member and publishes in conferences on machine learning. He was a recipient of an outstanding paper runner-up award at the 2022 International Conference on Machine Learning, or ICML, and an NSF CAREER Award in 2025.

Abstract:

Interactive imitation learning has emerged as a practical approach for solving complex control tasks, including for robotics and video games. By making interactive queries to a demonstration expert, a learning agent can collect targeted feedback to refine its policy, often achieving significantly higher sample efficiency than methods that rely solely on offline expert demonstrations.

In this talk, Zhang advances the theoretical understanding of the value of interaction in imitation learning from a sample efficiency perspective. In the realizable setting, he shows that when expert annotation cost is measured on a per-state basis, interactive imitation learning provides provable cost savings over offline approaches. In the presence of model misspecification, he further demonstrates that interactive methods can achieve stronger competitiveness guarantees relative to the expert policy than purely offline methods. Finally, motivated by practical implementations of imitation learning, he shows that combining offline datasets with interactive expert annotations can, in some cases, be substantially more cost-efficient than relying on either source alone. He also verifies theoretical findings with simulations on MuJoCo control tasks.

Seminar host:

Our host for this LIONS Seminar is Shaofeng Zou, an associate professor in the School of Electrical, Computer and Energy Engineering, part of the Ira A. Fulton Schools of Engineering at Arizona State University. In addition, he currently serves as an associate editor for IEEE Transactions on Signal Processing. His honors and awards include a 2020 NSF Computer and Information Science and Engineering Research Initiation Initiative Award and a 2024 NSF CAREER Award.

Visit the LIONS Seminar website for more information.

LIONS Seminar: Understanding the utility of interaction in imitation learning
Friday, March 6, 2026
1:30–2:30 p.m.
Goldwater Center for Science and Engineering, room 487 [map]
Attend on Zoom