Projects
Here are some of my past research projects!
- Customizable Robot Policies We introduced the Adjustable Control of Reinforcement Learning Dynamics (ACORD) algorithm, which lets users customize and adjust robot behavior in real-time.
(accompanying publication: "Online Behavior Modification for Expressive Control of RL-Trained Robots")
- Facilitate Novel User Interventions: We introduced the Imaginary Out-of-Distribution Actions (IODA) algorithm which allows users to leverage existing robot behaviors to accomplish new tasks (like watering flowers!).
(accompanying publication: "Imagining In-distribution States: How Predictable Robot Behavior Can Enable User Control Over Learned Policies")