Preprints: 'Data-efficient Kernel Methods for Learning Differential Equations and their Solution Operators: Algorithms and Error Analysis'; 'Data-efficient Kernel Methods for Learning Hamiltonian Systems'.
Research Experience
During her PhD, focused on developing efficient data-driven methods for learning and solving partial differential equations and their solution operators, and participated in several related research projects.
Education
PhD Candidate: Applied and Computational Mathematics at Caltech, advised by Professors Houman Owhadi and Franca Hoffmann; Bachelor's Degree: Mathematics and Computer Science from École Polytechnique.
Background
Research Interests: The intersection of mathematical analysis and machine learning, with a primary focus on the theoretical foundations of data-driven methods for complex systems. Particularly interested in kernel methods and Gaussian processes for learning and solving partial differential equations, with an emphasis on analyzing error bounds and function approximation properties across various model architectures and problem settings.