Cyrus Neary of University of British Columbia

How can we transform artificial intelligence (AI) and machine learning capabilities into reliable, autonomous robotic systems? How can we engineer AI systems within budget constraints, certify them with respect to stakeholder requirements, and ensure that they meet the needs of the end user? Answering these questions necessitates new engineering methodologies for AI systems, as well as AI algorithms that leverage the unique characteristics of engineering problems. In this talk, I will begin by presenting methods that integrate foundation models such as large language models and vision-language-action models with frameworks and algorithms for verifiable sequential decision-making. I will then present compositional approaches to reinforcement learning, which enable independent development and testing of separate learning-enabled modules and facilitate the reliable deployment of their compositions in practice. Finally, I will present control-oriented learning algorithms that combine data with prior physics knowledge, yielding learning-enabled systems that effectively control hardware after mere minutes of data collection and training. Experiments on robotic hardware, ranging from manipulators to ground vehicles to hexacopters, demonstrate the important role that these algorithms play in the fast and reliable transfer of learning-driven algorithms to their target, real-world operating environments.

Biography:

Cyrus Neary is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of British Columbia. His research develops theory and algorithms that enable scalable, generalizable, and modular artificial intelligence systems for autonomy, robotics, dynamics modeling, and control. Prior to joining UBC, he was a postdoctoral researcher at Mila – the Québec AI Institute. Cyrus received his Ph.D. and M.Sc. in Computational Science, Engineering, and Mathematics from the University of Texas at Austin and his B.A.Sc. in Engineering Physics from the University of British Columbia.

Email: