Want to receive email updates?

Wednesday, January 15

special colloquium: Reinforcement Learning and Psychology, Rich Sutton, 12pm Thursday Jan 23

Come hear one of the pioneers in the field!

Reinforcement Learning and Psychology: A Personal Story

1200pm-100pm Thursday, January 23
East Hall 4448

Rich Sutton
Reinforcement Learning and Artificial Intelligence Laboratory
University of Alberta


The modern field of reinforcement learning (RL) has a long, intertwined relationship with psychology. Almost all the powerful ideas of RL came originally from psychology, and today they are recognized as having significantly increased our ability to solve difficult engineering problems such as playing backgammon, flying helicopters, and optimal placement of internet advertisements. Psychology should celebrate this and take credit for it! RL has also begun to give something back to the study of natural minds, as RL algorithms are providing insights into classical conditioning, the neuroscience of brain reward systems, and the role of mental replay in thought. I have been working in the field of RL for much of this journey, back and forth between nature and engineering, and have played a role in some of the key steps. In this talk I tell the story as it seemed to happen from my point of view, summarizing it in four things that I think every psychologist should know about RL: 1) that it is a formalization of learning by trial and error, with engineering uses, 2) that it is a formalization of the propagation of reward predictions which closely matches behavioral and neuroscience data, 3) that it is a formalization of thought as learning from replayed experience that again matches data from natural systems, and 4) that there is a beautiful confluence of psychology, neuroscience, and computational theory on common ideas and elegant algorithms. 


Richard Sutton is a professor and iCORE chair in the department of computing science at the University of Alberta.  He is a fellow of the Association for the Advancement of Artificial Intelligence and co-author of the textbook Reinforcement Learning: An Introduction from MIT Press. Before joining the University of Alberta in 2003, he worked in industry at AT&T and GTE Labs, and in academia at the University of Massachusetts. He received a PhD in computer science from the University of Massachusetts in 1984 and a BA in psychology from Stanford University in 1978.  Rich's research interests center on the learning problems facing a decision-maker interacting with its environment, which he sees as central to artificial intelligence.  He is also interested in animal learning psychology, in connectionist networks, and generally in systems that continually improve their representations and models of the world.




--------------------------
Richard L. Lewis                                rickl@umich.edu
Professor                                          http://www-personal.umich.edu/~rickl/
Department of Psychology                 Voice:  (734) 763-1466
University of Michigan                       Fax:    (734) 647-9440
530 Church Street                             Office:  East Hall 3018
Ann Arbor, MI  48109-1043