Reinforcement learning in the real world - how to "cheat" and still feel good about it.

Speaker: Prof. Matthew E. Taylor from the University of Alberta

Abstract:
Reinforcement learning has had many successes, but significant amounts of time and/or data can be required to reach acceptable performance. If agents or robots are to be deployed in real-world environments, it is critical that our algorithms take advantage of existing data and human knowledge. This talk will discuss a selection of recent work that improves reinforcement learning by leveraging demonstrations and feedback from imperfect users, with an emphasis on how interactive machine learning can be extended to best leverage the unique abilities of both computers and humans.

Bio:
Matt is an Associate Professor of Computing Science at the University of Alberta and a Fellow and Fellow-in-Residence at Amii. He is the Director of the Intelligent Robot Learning (IRL) Lab and a Principal Investigator at the Reinforcement Learning & Artificial Intelligence (RLAI) Lab, both at the University of Alberta. Matt was formerly a Principal Researcher at Borealis AI in Edmonton, the artificial intelligence research lab for the Royal Bank of Canada. His current research interests include fundamental improvements to reinforcement learning, applying reinforcement learning to real-world problems, and human-agent interaction. He also is teaching a fall graduate course on Interactive Machine Learning, which is all freely accessible at https://sites.google.com/ualberta.ca/cmput656/.

Видео Reinforcement learning in the real world - how to "cheat" and still feel good about it. канала WaterlooAI