Mathematics of Deep Learning Seminar: Eric Darve

Date & Time

Title: Physics-informed machine learning: open mathematical questions

Abstract: The requirements and constraints to numerically solve physics-informed machine learning problems (PhysML) are notably different from classical machine learning for image and signal processing. We will discuss some of the current mathematical challenges that are impeding the wider adoption of these numerical methods in engineering. We will also discuss some of the challenges and research opportunities for the specific case of numerical optimizers in PhysML.

Advancing Research in Basic Science and MathematicsSubscribe to Flatiron Institute announcements and other foundation updates