Research Statement

I am a Computer Science Ph.D. student studying Reinforcement Learning, which is the problem of devising machines which are capable of autonomously learning to solve multi-stage decision problems. A multi-stage decision process is a problem which has multiple steps, where effective performance in the problem depends on the sequence of actions taken by an agent in the problem.

My particular interest is in investigating through mathematical and scientific means the learning algorithms which we have found to be effective in complex problems. My goal is to discover the mechanisms which enable the success of these methods and to leverage the underlying mathematical structures to devise better learning algorithms for the future.