William Pinstrup Huey

Hi, I’m Will. I’m a Senior at Cornell University studying Computer Science, and I’m currently working with Prof. Sanjiban Choudhury and the PoRTAL Lab on learning robotic control from video demonstrations.

In the past, I worked under Prof. Marco Hutter at the ETH Zurich Robotic Systems Lab and Prof. Don Greenberg at the Cornell Program of Computer Graphics. I spent a summer as an intern at NASA and another at Amazon Web Services. At Cornell, I’ve maintained a 4.1 GPA, I was a teaching assistant for CS 1620: Visual Imaging in the Electronic Age, and I’m a Rawlings Presidential Research Scholar.

I also enjoy running marathons, climbing walls, and playing foot volley.

News

Mar 04, 2025	I won $500 and first place at the RCPRS Grand Slam for my presentation on learning from videos
Feb 24, 2025	I was a panelist at ACSU research night, where I had the opportunity to share my awesome experience doing research at Cornell with other undergraduates
Oct 30, 2024	Time Your Rewards was accepted to the 2024 CoRL workshop on Whole Body Control and Bimanual Manipulation
Oct 26, 2024	I presented my work with Sean Brynjólfsson on visual traversability prediction to prospective students at the Cornell Discover Undergraduate Research in Engineering Showcase.

Selected Publications

Imitation Learning from a Single Temporally Misaligned Video

William Huey, Huaxiaoyue Wang, Anne Wu, Yoav Artzi, and Sanjiban Choudhury

In review, 2025

TLDR arXiv Code

To learn from a single video, current approaches leverage an image encoder, distance metric on the encoder’s latent space, and distribution-matching algorithm to produce dense rewards for RL. We prove that frame-level distribution matching algorithms cannot accurately model the ground truth reward of a video demonstration. We propose ORCA (ORdered Coverage Alignment), an intuitive algorithm that matches distributions at the sequence level, achieving better performance on 16 video-imitation tasks in two simulation environments.