David S. Hayden

Summary

I lead Perception AI Research at Cruise. My work in machine learning and computer vision has shipped to hundreds of driverless cars, ran live in stadiums of 40,000 people, supported seed and Series A rounds, and is published in top conferences and journals including ICML, CVPR, Neurips, and Nature.

Research

I work on generative and world models, foundation model alignment and guidance, longtail robustness, uncertainty quantification, and synthetic data. I received my PhD at MIT working on interpretable machine learning and computer vision with focus on behavior analysis, multi-object tracking, Bayesian nonparametrics for time-series, distributions on manifolds, and uncertainty to guide decision making.

News

My paper Generative Data Mining with Longtail-Guided Diffusion was accepted at ICML '25!
Our paper DriveGPT: Scaling Autoregressive Behavior Models for Driving was accepted at ICML '25!
I was asked to lead Perception AI Research at Cruise in January, 2025.
Our paper Causal Composition Diffusion Model for Closed-loop Traffic Generation was accepted at CVPR '25!
I'm helping organize RSS 2024's Workshop on Autonomous Vehicles Across Scales and organizing an afterparty. Come join us in Delft!
My real-time, camera-based vision system successfully tracked all championship barrel racers and computed performance statistics live at the 2023 American Rodeo. Featured on the jumbotron to an audience of more than 40,000 and aired on TV at CBS Sports)! Example video.

Publications
Select Papers Conference Journal Book Chapter Workshop

Generative Data Mining with Longtail-Guided Diffusion
David S Hayden, M Ye, T Garipov, G Meyer, C Vondrick, Z Chen, Y Chai, E Wolff, S Srinivasa
ICML 2025
Paper

DriveGPT: Scaling Autoregressive Behavior Models for Driving
X Huang, E Wolff, P Vernaza, T Phan-Minh, H Chen, David S Hayden, et al.
ICML 2025
Paper

Causal Composition Diffusion Model for Closed-loop Traffic Generation
H Lin, X Huang, T Phan-Minh, David S Hayden, H Zhang, D Zhao, S Srinivasa, E Wolff, H Chen
CVPR 2025
Paper

Uncertainty Quantification and Structure Discovery for Scalable Behavior Science
David S. Hayden (Advisor: John W. Fisher III, Comittee: Robert Desimone, Justin Solomon)
PhD Thesis 2021
Thesis

Sequential Bayesian Experimental Design with Variable Cost Structure
Sue Zheng, David S. Hayden, Jason Pacheco, John W. Fisher III
Neurips 2020
Paper Supplemental

Efficient Data Association and Uncertainty Quantification for Multi-Object Tracking
David S. Hayden, Sue Zheng, John W. Fisher III
arXiv 2020
Paper Code

Nonparametric Object and Parts Modeling with Lie Group Dynamics
David S. Hayden, Jason Pacheco, John W. Fisher III
CVPR 2020
Paper Supplemental Video Code

Atypical Behaviour and Connectivity in SHANK3-Mutant Macaques
Yang Zhou, Jitendra Sharma, Qiong Ke, Rogier Landman, Jingli Yuan, Hong Chen, David S. Hayden et al.
Nature, June 2019
Paper Code

Unobtrusive, Wearable Social Interaction Detection and Assistance
David S. Hayden, Robert C. Miller, Seth Teller
CHI 2014 Workshop on Assistive Augmentation
Paper

The Accuracy-Obtrusiveness Tradeoff for Wearable Vision Platforms
David S. Hayden, Carl Vondrick, Stella Jia, Yafim Landa, Robert C. Miller, Antonio Torralba, Seth Teller
CVPR 2012 Workshop on Egocentric Vision
Paper