Robot Learning with Minimal Human Feedback

Explore a cutting-edge robotics colloquium featuring Erdem Biyik from USC, focusing on robot learning with minimal human feedback. Delve into innovative techniques that enable training robots using limited human input, such as single demonstrations, language instructions, or natural eye gaze. Discover how reinforcement learning from human feedback can be enhanced through language corrections to improve data efficiency. Learn about the application of large pretrained vision-language models in generating direct supervision for robot learning. Gain insights into overcoming the challenge of limited robotics datasets and the potential for breakthroughs in the field. Understand the speaker's background, including his role as an assistant professor at USC, his postdoctoral work at UC Berkeley, and his academic journey through Stanford University and Bilkent University.