AI Seminar

Rethinking Rewards in Reinforcement Learning

Satinder Singh

Professor, Department of EECS
University of Michigan
Tuesday, November 24, 2009
4:00pm - 5:30pm
Stained-Glass Conference Room (3725 Beyster Bldg.)

Add to Google Calendar

About the Event

In the computational reinforcement learning (RL) framework, rewardsómore specifically, reward functionsódetermine the problem the learning agent is trying to solve. Properties of the reward function influence how easy or hard the problem is, and how well an agent may do, but RL theory and algorithms are completely insensitive to the source of rewards. This is a strength of the framework because of the generality it confers, but it is also a weakness because it defers key questions about the nature of reward functions. In this talk, I address this weakness from two directions. First, I consider the role of evolution in determining where rewards come from in natural agents. Specifically, I present a computational framework in which evolved rewards capture regularities across environments leaving the agent to learn regularities within its environment during its lifetime. Second, I describe how in designing artificial agents the current use of rewards confounds their role in defining preferences over behaviors and their role as parameters of actual agent behavior (RL agents act so as to maximize reward). Disentangling this "preferences parameters confound" can be beneficial in designing artificial agents. I will present many empirical illustrations of both of these aspects of rethinking rewards in RL. This talk describes joint work with Richard Lewis, Andrew G. Barto, Jonathan Sorg and Akram Helou.

Additional Information

Sponsor(s): Toyota

Open to: Public