Associate Professor

University of Pennsylvania

I am an associate professor at UPenn’s GRASP lab, with a primary appointment in CIS, and a secondary appointment in ESE. I lead the Perception, Action, and Learning (PennPAL) Research Group, where we work on problems at the intersection of robotics, machine learning, and computer vision.

Research Statement: My research statement, last updated Aug 2025, explains our research vision.

Teaching: In Fall 2025, I am teaching CIS 4190 / 5190 Applied Machine Learning. Here are the past courses I have taught.

For prospective masters and undergraduate studentsIf you're an undergraduate or MS student interested in collaborating with our PennPAL research group, please complete this form. We monitor responses regularly, and will reach out to you if there's an opening.
For prospective PhD studentsFormal admissions are processed through the university's various programs. You may send me email too, but please note that I might not be able to respond.
Emailing meI receive a lot of email and find it difficult to respond to them all. I will try to get to your email sooner if you are enrolled at Penn and either: (1) a student in a class I am instructing (most communication would be appropriate on the Ed forum, unless you specifically want to avoid TA eyes; if you're sure you need to email me, start subject with "[<course number> student]", e.g. "[CIS 4190 student]"), or (2) assigned to me as an advisee (start email subject with "[Student Advisee]").
Recent Publications
Motion Capture with Millimeter-Wave Tags
. Motion Capture with Millimeter-Wave Tags. SenSys, 2026.
Let it Cook: Learning to Wait in Sequential Decision Making
. Let it Cook: Learning to Wait in Sequential Decision Making. Reinforcement Learning Conference (RLC), 2026.
TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
. TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance. ICML, 2026.
🏆 ICML 2026 Spotlight
Expanding Spatial and Temporal Context for Robotic Imitation Learning Policies With Scene Graphs
. Expanding Spatial and Temporal Context for Robotic Imitation Learning Policies With Scene Graphs. CVPR, 2026.
Correspondence-Driven Trajectory Warping for Data-Efficient Imitation and Autonomous Play
. Correspondence-Driven Trajectory Warping for Data-Efficient Imitation and Autonomous Play. ICLR, 2026.

Funding & Support

Our work is possible thanks to the support of: