Bio. I am a fourth-year Ph.D. student in Statistics at Stanford University advised by John Duchi. My research interests lie in a convex combination of machine learning, statstics and optimization, and especially in how to quantify models uncertainty and make them more robust to changing environments.

Previously, I completed my undergraduate studies at Ecole Polytechnique from which I obtained a B.S. and a M.S. in 2016 and 2017. I also spent internships at Bloomberg LP as a Quantitative Researcher in 2017, where I worked with Bruno Dupire and Julien Guyon, and at Google as a Data Scientist in 2019.



Robust Validation: Confident Predictions Even When Distributions Shift
Maxime Cauchois, Suyash Gupta, Alnur Ali, John Duchi.
In Submission
[arxiv] [pdf]
Knowing what you know: valid confidence sets in multiclass and multilabel prediction.
Maxime Cauchois, Suyash Gupta, John Duchi.
Accepted to the Journal of Machine Learning Research, March 2021
[arxiv] [pdf]



  • STATS 100: Mathematics of Sports (Fall 2019).
  • Teaching Assistant

  • STATS 361: Causal Inference (Spring 2020).
  • EE 364A: Convex Optimization (Winter 2020).
  • STATS 322: Gaussian estimation: Sequence and wavelet models (Fall 2019).
  • STATS 310A: Theory of Probability I (Fall 2018).
  • STATS 101: Introduction to Data Science (Summer 2018).
  • STATS 116: Introduction to Probability (Fall 2017).