Machines today excel at seemingly complex games such as chess and Jeopardy, yet still struggle with basic perceptual, planning, and motor tasks in the physical world. What are the appropriate representations needed to execute and adapt robust behaviors in real-time? I will present some examples of learning algorithms from my group that have been applied to robots for monocular visual odometry, high-dimensional trajectory planning, and legged locomotion. These algorithms employ a variety of techniques central to machine learning: dimensionality reduction, online learning, and reinforcement learning. I will show and discuss applications of these algorithms to autonomous vehicles and humanoid robots.
Daniel Lee is the Evan C Thompson Term Chair, Raymond S. Markowitz Faculty Fellow, and Professor in the School of Engineering and Applied Science at the University of Pennsylvania. He received his B.A. summa cum laude in Physics from Harvard University in 1990 and his Ph.D. in Condensed Matter Physics from the Massachusetts Institute of Technology in 1995. Before coming to Penn, he was a researcher at AT&T and Lucent Bell Laboratories in the Theoretical Physics and Biological Computation departments. He is a Fellow of the IEEE and has received the National Science Foundation CAREER award and the University of Pennsylvania Lindback award for distinguished teaching. He was also a fellow of the Hebrew University Institute of Advanced Studies in Jerusalem, an affiliate of the Korea Advanced Institute of Science and Technology, and organized the US-Japan National Academy of Engineering Frontiers of Engineering symposium. As director of the GRASP Robotics Laboratory and co-director of the CMU-Penn University Transportation Center, his group focuses on understanding general computational principles in biological systems, and on applying that knowledge to build autonomous systems.
