The field of robotics has recently witnessed a significant influx of learning-based methodologies, revolutionizing areas such as manipulation, navigation, locomotion, and drone technology. This talk aims to delve into the forefront of robot learning systems, particularly focusing on their scalability and deployability to open-world problems, through two main paradigms of learning-based methods for robotics: reinforcement learning and imitation learning.

Bio:

Zipeng Fu is a CS PhD student at Stanford AI Lab, advised by Chelsea Finn. His research focuses on deployable robot systems and learning in the unstructured open world. His representative work includes Mobile ALOHA, Robot Parkour Learning, and RMA, receiving CoRL 2023 & 2022 Best System Finalist awards. His research is supported by Stanford Graduate Fellowship as a Pierre and Christine Lamond Fellow. Previously, he was a student researcher at Google DeepMind. He completed his master's at CMU and bachelor’s at UCLA.

Homepage: zipengfu.github.io/