Research
Publications
Batlle, P., Darcy, M., Hosseini, B., & Owhadi, H. (2023). Kernel Methods are Competitive for Operator Learning (Journal, Arxiv)
Darcy, M., Hamzi, B., Livieri, G., Owhadi, H., & Tavallali, P. (2023). One-shot learning of stochastic differential equations with data adapted kernels. Physica D: Nonlinear Phenomena, 444, 133583. (Journal, Arxiv).
Preprints
M. Darcy, B. Hamzi,, J. Susiluoto, A. Braverman, H. Owhad. (2021). Learning dynamical systems from data: a simple cross-validation perspective, part II: nonparametric kernel flows. (researchgate)
Talks and presentations
Talk: "Kernel methods and PINNs for rough partial differential equations" Digital Twins for inverse problems in Earth sciences (CIRM) - July 22nd, 2024.
Talk: "Kernel methods and PINNs for rough partial differential equations" SciCADE, National University of Singapore, Singapore, July 15th, 2024.
Talk: Research overview, Franca Hoffman group meeting - May 31st, 2024.
Talk: "Kernel methods for rough partial differential equations" Southern California Applied Mathematics Workshop, UC San Diego, USA, April 27th 2024.
Talk: "Kernel methods for rough partial differential equations" SIAM Material sciences - May 20th, 2024.
Talk: "Kernel methods for rough partial differential equations" SIAM Uncertainty Quantification 2024 - March 1st 2024 (slides).
Talk: "One shot learning of SDEs with Gaussian processes" at the International Conference: Differential Equations for Data Science - February 22nd 2024 (slides).
Talk “Kernel methods for operator learning” at the One World Mathematics of Machine Learning seminar (online), October 18th, 2023 (watch).
Talk: “Kernel Methods are Competitive for Operator Learning” at the 2nd IACM Mechanistic Machine Learning and Digital Engineering for Computational Science Engineering and Technology, University of El Paso, Texas, USA - September 23rd 2023.
Talk: "Kernel Methods are Competitive for Operator Learning" at 10th International Congress on Industrial and Applied Mathematics - August 22nd 2023 (slides).
Talk: "Kernel Methods are Competitive for Operator Learning" at the Argonne National Lab LANS Seminar - August 2nd 2023 (slides).
Talk: "One Shot Learning of Stochastic Differential Equations with Gaussian Processes" at the SIAM conference on the Application of Dynamical Systems- May 18th 2023 (slides).
Talk: "Kernel methods are competitive for Operator Learning", DataSig Rough Path Interest Group (online)- May 11th 2023.
Talk: "Benchmarking Operator Learning with Simple and interpretable Kernel Methods" at the Workshop on Establishing Benchmarks for Data-Driven Modeling of Physical System - April 6th, 2023 (slides).
Talk: "One shot learning of stochastic differential equations with Gaussian Processes" - SIAM Computational Sciences and Engineering, Amsterdam - March 6th, 2023.
Poster - "One shot learning of stochastic differential equations with Gaussian Processes" - Third Symposium on Dynamical Systems and Machine Learning 2022, Fields Institute, Toronto, CA (poster).
Talk - "One shot learning of stochastic differential equations with Gaussian processes and computational graph completion" - ICCOPT 2022, Lehigh, PA, USA (slides).
Talk - "Kernel Flows Demystified: Application to Regression" - Second Symposium on Dynamical Systems and Machine Learning 2020, Fields Institute, Toronto, CA (watch).
mdarcy@ caltech.edu