Interests: Reinforcement Learning

My research goal is to develop intelligent sequential-decision making agents that require minimal human intervention to achieve goals. To this end, most of my research is in reinforcement learning (RL). I have worked in deep RL, inverse RL, and goal-conditioned RL. More recently, I have begun to use game-theoretic online learning tools and have also investigated temporal abstraction for exploration.

Projects | Talks


  • Revisiting Overestimation in Value-based Deep Reinforcement Learning | AMII AI Seminar (Slides) (Video)
  • Trajectory-Ranked Reward Extrapolation | Internal talk at Preferred Networks (Slides)
  • Deterministic Implementations for Reproducibility in Deep Reinforcement Learning | AAAI 2019 Workshop on Reproducibility in AI (Slides)


Deep Reinforcement Learning

My Master's thesis focused on reproducibility in deep reinforcement learning. Specifically, I studied the impact of nondeterminism in algorithm implementations on our ability to reproduce results.


At Preferred Networks I have worked on targeted grasping for robotics. We combine techniques from goal-conditioned reinforcement learning (hindsight experience replay), deep reinforcement learning (QT-OPT), and distributed training to achieve target grasping on a human support robot.

Inverse Reinforcement Learning

We leverage ranked demonstrations to improve upon a suboptimal demonstrator in high-dimensional deep reinforcement learning tasks.