Jonathan Deaton

San Mateo, California

I’m a senior software engineer and machine learning researcher at X broadly interested in probabilistic ML.

I believe robust ML systems require theoretical foundations in probabilistic reasoning and statistics. I enjoy translating cutting-edge ML methods into practical applications.

My current interests include Equivariant Euclidian GNNs for molecular modeling, generative language models in biology, and diffusion/score-based generative modeling.

I have a background in Bioengineering and Computer Science at Stanford where I also completed premedical requirements.

I was a gymnast for 15 years, competing for Stanford Mens Gymnastics and represented the Unites States on the Junior National Team at competitions in China, England, and Colombia.

news

Nov 2, 2021	Joined Google X! Applied ML research in computational biology.
Sep 2, 2019	Joined Google Health working on AI in computer vision for medicine.
Mar 16, 2018	Starting Masters in Computer Science at Stanford!

selected publications

JMLR

Underspecification presents challenges for credibility in modern machine learning

Alexander DAmour, Katherine Heller, Dan Moldovan, and 34 more authors

Journal of Machine Learning Research, 2020

Bib

@article{Underspecification,
  title = {Underspecification presents challenges for credibility in modern machine learning},
  author = {DAmour, Alexander and Heller, Katherine and Moldovan, Dan and Adlam, Ben and Alipanahi, Babak and Beutel, Alex and Chen, Christina and Deaton, Jonathan and Eisenstein, Jacob and Hoffman, Matthew D and Hormozdiari, Farhad and Houlsby, Neil and Hou, Shaobo and Jerfel, Ghassen and Karthikesalingam, Alan and Lucic, Mario and Ma, Yian and McLean, Cory and Mincu, Diana and Mitani, Akinori and Montanari, Andrea and Zachary Nado, Vivek Natarajan and Nielson, Christopher and Osborne, Thomas F and Raman, Rajiv and Ramasamy, Kim and Rory Sayres, Jessica Schrouff and Seneviratne, Martin and Sequeira, Shannon and Suresh, Harini and Veitch, Victor and Max Vladymyrov, Xuezhi Wang and Webster, Kellie and Yadlowsky, Steve and Yun, Taedong and Zhai, Xiaohua and Sculley, D},
  journal = {Journal of Machine Learning Research},
  year = {2020},
}

Addressing the real-world class imbalance problem in dermatology

Wei-Hung Weng, Jonathan Deaton, Vivek Natarajan, and 2 more authors

Proceedings of the Machine Learning for Health NeurIPS Workshop, PMLR, 2020
Mini‐Metagenomics and Nucleotide Composition Aid the Identification and Host Association of Novel Bacteriophage Sequences

Jonathan Deaton, Feiqiao Brian Yu, and Stephen R Quake

Advanced biosystems, 2019