This seminar will only take place online. Please contact [email protected] for the Zoom link.

Speaker: Carlos Fernandez-Granda, Courant Institute of Mathematical Science and the Center for Data Science, New York University

Title: Deep learning for scientific discovery

Abstract: Deep-learning models for image processing achieve impressive results when trained on standard natural-image datasets in a supervised fashion. However, unleashing their potential in practice will require developing unsupervised or semi-supervised approaches capable of learning from real data, as well as understanding the strategies learned by these models. In this talk, we will describe advances in this direction motivated by a real-world scientific application: determining the 3D atomic structure of catalytic nanoparticles from extremely noisy electron-microscope data.

Short bio: Carlos Fernandez-Granda is an Associate Professor of Mathematics and Data Science at the Courant Institute of Mathematical Science and the Center for Data Science in New York University. The goal of his research group is to design and analyze data-science methodology, motivated by applications in medicine, climate science and scientific imaging.