Abstract
Recent advances in machine learning have transformed multiple AI-related fields. Notably, robust general-purpose language and vision models are fast becoming reality and these new capabilities have already begun making their way into consumer-facing technologies where they affect the lives of many millions of people. These same underlying advancements also portend sea-change in robotics. It is now possible to reliably imbue robots with new behaviors, such as beating eggs or folding clothing, using just an hour or two of teaching and a few dozen GPU-hours of compute. In this talk, I will discuss our team's push, at the Toyota Research Institute, to scale ML-powered robot behavior teaching and the road ahead to general-purpose Large Behavior Models for robots. These models will possess the flexibility and generality of existing Large Language Models, but will be capable of dexterously controlling a robot to effect change in the physical world.
Bio
Ben Burchfiel is manager and co-lead for the Large Behavior Model project at Toyota Research Institute where he leads a team working to create general-purpose robots via machine learning at scale. Before joining TRI, Ben Burchfiel was a Postdoc at Brown University and a PhD student at Duke University in the Intelligent Robot Lab. During his thesis work, Ben focused on 3D perception, learning from demonstration, and multimodal vision-language representations. Prior to that, Ben received his Bachelor of Science from UW-Madison, where he studied Computer Science. These days, Ben's research focuses on making general-purpose robots a reality by imbuing data-driven methods with soft inductive biases that make minimal assumptions about the structure of the world and relax gracefully with scale.
21 авг 2024
