Mathematical Foundations of Deep Learning (364-0-31)

Instructors

Benjamin Weinkove
847 4915587
Lunt Hall 310 2033 Sheridan Road

Meeting Info

Lunt Hall 107: Mon, Wed, Fri 9:00AM - 9:50AM

Overview of class

MATH 364-0 Mathematical Foundations of Deep Learning (1 Unit) Neural networks and analysis of the approximation properties which underlie the mathematical principles of deep learning. Loss functions and training phenomena. Comparison of different activation functions and network architectures. Vapnik-Chervonenkis dimension and bit extraction.
Prerequisites: MATH 320-1 or MATH 321-1.

Class Materials (Required)

Mathematical Theory of Deep Learning
by Peterson and Zech
https://arxiv.org/pdf/2407.18384

Class Materials (Suggested)

No suggested materials.

Enrollment Requirements

Enrollment Requirements: Prerequisites: MATH 320-1 or MATH 321-1.

Associated Classes

DIS - Lunt Hall 104: Thurs 9:00AM - 9:50AM

DIS - Lunt Hall 103: Thurs 9:00AM - 9:50AM