Presentation Type
Oral
Student
No
Abstract
In recent years machine learning, and in particular deep learning has emerged as a powerful and robust tool for solving problems in fields ranging from robotics, to medicine, materials science, cosmology and beyond. As work on applications has advanced, so too has theory advanced to guide, explain, interpret deep learning. In this talk I will provide an overview on universality of neural networks. I will explain what it means for a neural network to be a universal approximator, qualitatively and quantitatively, and give examples of these results applied to existing networks. Finally, I will conclude by discussing some recent work on manifold universality in the nascent field of geometric machine learning.
Start Date
2-7-2023 9:50 AM
End Date
2-7-2023 10:50 AM
Session 4: An Overview of Deep Learning and Universality
Pasque 255
In recent years machine learning, and in particular deep learning has emerged as a powerful and robust tool for solving problems in fields ranging from robotics, to medicine, materials science, cosmology and beyond. As work on applications has advanced, so too has theory advanced to guide, explain, interpret deep learning. In this talk I will provide an overview on universality of neural networks. I will explain what it means for a neural network to be a universal approximator, qualitatively and quantitatively, and give examples of these results applied to existing networks. Finally, I will conclude by discussing some recent work on manifold universality in the nascent field of geometric machine learning.