Aman Sinha

I am a co-founder of Trustworthy AI, which was acquired by Waymo in 2021. At Waymo, I continue Trustworthy AI's mission to build an automated test generation and risk modeling platform for safety-critical software. My work spans the intersection of machine learning, optimization, statistics, and control theory. Overall, I try to make computers faster and smarter on problems that matter to people.

Previously, I received a PhD at Stanford, where my advisor was John Duchi. My main research focused on distributionally robust optimization and rare-event simulation, and it involved many collaborations with Russ Tedrake at MIT. I was a member of the Machine Learning Group, and I also researched medical applications of machine learning as part of the Wearable Health Lab. I was supported by a Stanford Graduate Fellowship and a Hertz Fellowship.

Before Stanford, I received an MPhil in Information Engineering at the University of Cambridge on a Churchill Scholarship, where I was advised by Glenn Vinnicombe and Carl Rasmussen. I received a BSE in Mechanical and Aerospace Engineering at Princeton; my thesis advisor was Naomi Leonard, and I also worked in the research groups of Howard Stone and Lex Smits.

I have been fortunate to work at and collaborate with companies including Toyota Research Institute (TRI), Quantifind, Microsoft, and Merck.

aman at trustworthy dot ai, CV, Linkedin, Google scholar

Rate-informed discovery via Bayesian adaptive multifidelity sampling.
Aman Sinha*, Payam Nikdel*, Supratik Paul, Shimon Whiteson. CoRL 2024.
[link] [pdf] [arxiv]

Embedding synthetic off-policy experience for autonomous driving via zero-shot curricula.
Eli Bronstein*, Sirish Srinivasan*, Supratik Paul*, Aman Sinha, Matthew O'Kelly, Payam Nikdel, Shimon Whiteson. CoRL 2022.
Oral presentation.
[link] [pdf] [arxiv]

Neural bridge sampling for evaluating safety-critical autonomous systems.
Aman Sinha*, Matthew O'Kelly*, Russ Tedrake, John Duchi. NeurIPS 2020.
[link] [pdf] [arxiv] [code] [poster] [slides] [talk] [press]

FormulaZero: distributionally robust online adaptation via offline population synthesis.
Aman Sinha*, Matthew O'Kelly*, Hongrui Zheng*, Rahul Mangharam, John Duchi, Russ Tedrake. ICML 2020.
[link] [pdf] [arxiv] [code] [slides] [talk]

Efficient black-box assessment of autonomous vehicle safety.
Justin Norden*, Matthew O'Kelly*, Aman Sinha*. NeurIPS 2019 Workshop on Machine Learning for Autonomous Driving, CVPR 2020 Workshop on Scalability in Autonomous Driving.
[arxiv] [blog]

Digital biomarkers of spine and musculoskeletal disease from accelerometers: Defining phenotypes of free-living physical activity in knee osteoarthritis and lumbar spinal stenosis.
Christy Tomkins-Lane, Justin Norden, Aman Sinha, Richard Hu, Matthew Smuck. The Spine Journal, 2019.
Outstanding Paper Award.
[link]

Scalable end-to-end autonomous vehicle testing via rare-event simulation.
Matthew O'Kelly*, Aman Sinha*, Hongseok Namkoong*, John Duchi, Russ Tedrake. NeurIPS 2018.
[link] [pdf] [arxiv] [code] [poster] [videos]

Certifying some distributional robustness with principled adversarial training.
Aman Sinha*, Hongseok Namkoong*, John Duchi. ICLR 2018.
Oral presentation.
[link] [pdf] [arxiv] [code] [slides] [talk]

Objective measurement of function following lumbar spinal stenosis decompression reveals improved functional capacity with stagnant real-life physical activity.
Matthew Smuck, Amir Muaremi, Patricia Zheng, Justin Norden, Aman Sinha, Richard Hu, Christy Tomkins-Lane. The Spine Journal, 2018.
Outstanding Paper Award.
[link]

Adaptive sampling probabilities for non-smooth optimization.
Hongseok Namkoong, Aman Sinha, Steve Yadlowsky, John Duchi. ICML 2017.
[link] [pdf] [code] [poster]

Objective measurement of free-living physical activity (performance) in lumbar spinal stenosis: are physical activity guidelines being met?
Justin Norden, Matthew Smuck, Aman Sinha, Richard Hu, Christy Tomkins-Lane. The Spine Journal, 2017.
Outstanding Paper Award Runner-up.
[link]

Learning kernels with random features.
Aman Sinha, John Duchi. NIPS 2016.
[link] [pdf] [code] [poster]

Dynamic management of network risk from epidemic phenomena.
Aman Sinha, John Duchi, Nick Bambos. IEEE CDC 2015.
[link] [pdf] [slides]

Visualizing the very-large-scale motions in turbulent pipe flow.
Leo Hellström, Aman Sinha, Lex Smits. Physics of Fluids, 2011.
[link] [pdf]

Past and present research collaborators (alphabetically by last name in reverse chronological order):
Payam Nikdel, Supratik Paul, Shimon Whiteson, Eli Bronstein, Matthew O'Kelly, Sirish Srinivasan, John Duchi, Russ Tedrake, Rahul Mangharam, Hongrui Zheng, Justin Norden, Richard Hu, Matthew Smuck, Christy Tomkins-Lane, Hongseok Namkoong, Amir Muaremi, Patricia Zheng, Steve Yadlowsky, Nick Bambos, Leo Hellström, Lex Smits

Safety-critical machine learning: development and testing.
Aman Sinha. Stanford University PhD. Thesis, 2020.
[link] [pdf] [slides] [talk]

Distributed gaussian process regression in networked systems.
Aman Sinha. University of Cambridge MPhil. Thesis, 2014.
[link] [pdf]

Distributed consensus protocols in adaptive multi-agent systems.
Aman Sinha. Princeton University Undergraduate Thesis, 2013.
Awarded Morgan W. McKinzie '93 Senior Thesis Prize for best senior thesis.
[link] [pdf]

Single-particle motion in colloids: nonlinear fluctuations in the presence of hydrodynamic interactions.
Aman Sinha. Princeton University Junior-Year Independent Study, 2012.
[link]

One time I had the honor of speaking to many people. [video] [text]