Assistant Professor | Harvard University
Samuel Gershman received his B.A. in Neuroscience and Behavior from Columbia University in 2007 and his Ph.D. in Psychology and Neuroscience from Princeton University in 2013. From 2013-2015 he was a postdoctoral fellow in the Department of Brain and Cognitive Sciences at MIT. He is currently an Assistant Professor in the Department of Psychology and Center for Brain Science at Harvard University.
Video games have recently become a popular platform to test reinforcement learning algorithms, which sometimes achieve human-level performance. But do these algorithms really learn and understand the world the way people do? Experiments with humans and machines suggest that the answer is no. This talk argues that building human-like algorithms requires a form of “theory-based” reinforcement learning.