Qinghua Liu

Research

Qinghua Liu studies machine learning for decision-making. His past research encompasses a wide range of areas within reinforcement learning, including multi-agent reinforcement learning, partially observable reinforcement learning, and reinforcement learning with large state spaces. Currently, he is particularly interested in reinforcement learning from human feedback and the development of foundation models for decision-making.

Biography

Qinghua Liu will join the Department of Computer Science at Washington University in St. Louis in 2025. He is currently a postdoctoral researcher at Microsoft Research in New York. He earned his Ph.D. from Princeton University in 2024 and his B.E. from Tsinghua University in 2018. He has also spent time at DeepMind in London and the Simons Institute. His work has been recognized with the Princeton SEAS Award and a Best Paper Award at the ICLR 2022 MARL workshop.