I'm currently at Tesla. I'm interested in probabilistic machine learning and computer vision.
| tonyduan (at) cs.stanford.edu | |
| @tonyduan_ |
| 2020 - Now | Machine Learning, Tesla Autopilot |
| 2019 - 2020 | Residency, Microsoft Research |
| 2017 - 2019 | MS in CS, Stanford |
| 2013 - 2017 | BS in EECS, UC Berkeley |
Diffusion Models from Scratch (2023). [HTML] [Code]
Includes complete derivations summarizing a few dozen of the most relevant papers.
Randomized Smoothing of All Shapes and Sizes
Greg Yang*, Tony Duan*, Edward J. Hu, Hadi Salman, Ilya Razenshteyn, Jerry Li
International Conference on Machine Learning (ICML), 2020 [Paper] [Code]
NGBoost: Natural Gradient Boosting for Probabilistic Regression
Tony Duan*, Anand Avati*, Daisy Ding, Khanh Thai, Sanjay Basu, Andrew Ng, Alejandro Schuler
International Conference on Machine Learning (ICML), 2020 [Paper] [Code]
Countdown Regression: Sharp and Calibrated Survival Predictions
Anand Avati, Tony Duan, Sharon Zhou, Kenneth Jung, Nigam Shah, Andrew Ng
Uncertainty in Artificial Intelligence (UAI), 2019 [Paper] [Code]
* Equal contribution.
I've reviewed for NeurIPS, ICML, ICLR.
Vision-based ML model for lane connectivity in autonomous or semi-autonomous driving
Patrick Cho, Ethan Knight, Tony Duan, Alex Xiao, Jason Lee
Tesla [Patent] [AI Day 2022]
Vision-based ML model for aggregation of static objects and systems for autonomous driving
Micael Carvalho, John Emmons, Patrick Cho, Bradley Emi, Saachi Jain, Nishant Desai, Tony Duan
Tesla [Patent] [AI Day 2021]
I've had opportunities to TA the following courses:
CS 229 (Machine Learning) at Stanford: Fall 2018.
CS 221 (Artificial Intelligence) at Stanford: Fall 2017.
CS 170 (Algorithms) at UC Berkeley: Fall 2016 and Spring 2017.