Abstract: Google Brain Robotics vision is to leverage learning to push the field of robotics forward. As such, we have engaged in research ranging in application from navigation to grasping and approach from deep RL to learning from demonstration. Fundamentally, our research is built around the core idea of lifelong learning. Our long term goal is to enable robots to continuously learn new tasks and refine existing ones. In this talk I will present a broad overview of our team’s research efforts followed by a deep dive into recent work on cognitive planning for navigation and a neuroscience inspired approach to learn connectivity of neural networks.

Bio: As a member of Google Brain Robotics, James Davidson’s research efforts are directed to robotic learning. In particular, James is investigating lifelong mobile manipulation, which spans from robot navigation and grasping to deep reinforcement learning (RL). Before joining Google in 2012, James worked at MITRE as a lead AI researcher and prior to that at Sandia National Laboratories. James received his Ph.D. from the University of Illinois in 2012 specializing in RL applied to robotics.

Homepage: https://research.google.com/teams/brain/robotics/