Ricardo Baptista

Postdoctoral Scientist, Amazon Search

avatar.jpg

I am a Postdoctoral Scientist at Amazon and a Visitor at Caltech hosted by Andrew Stuart and Houman Owhadi. My research is on probabilistic modeling and inference for problems in science and engineering. Most recently, I have been developing and analyzing generative models based on computational measure transport.

In Fall 2025, I am excited to join the University of Toronto as an Assistant Professor in Statistical Sciences and Faculty Affiliate at the Vector Institute!

Bio: From 2022-2024 I was a von Kármán instructor at Caltech in Computing + Mathematical Sciences. I completed my PhD at MIT in Computational Science and Engineering where I was fortunate to be advised by Youssef Marzouk. A copy of my thesis can be found here. Before MIT, I received my BASc in Engineering Science from the University of Toronto.

Contact: rsb (at) caltech (dot) edu
Follow: Google Scholar arXiv GitHub LinkedIn

Announcements

Sep 2024 Our paper on approximation theory for measure transport algorithms was accepted in AMS: Mathematics of Computations!
Jul 2024 Gave a joint keynote talk at the CIRM-Marseille Digital Twins for Inverse Problems Workshop on Bayesian inference via dimension reduction. Thank you to all of the organizers!
Nov 2023 Gave the USNCCM Large-Scale TTA early-career colloquium on dimension reduction methods for probabilistic modeling. Thank you Shelly and Patrick for the invitation!

Selected publications

  1. SIAM Review
    Coupling techniques for nonlinear ensemble filtering
    Alessio Spantini, Ricardo Baptista, and Youssef Marzouk
    SIAM Review, 2022
  2. FoCM
    On the representation and learning of monotone triangular transport maps
    Ricardo Baptista, Youssef Marzouk, and Olivier Zahm
    Foundations of Computational Mathematics, 2023
  3. JUQ
    Conditional Sampling with Monotone GANs: from Generative Models to Likelihood-Free Inference
    Ricardo Baptista, Bamdad Hosseini, Nikola B Kovachki, and Youssef Marzouk
    SIAM/ASA Journal on Uncertainty Quantification, 2024