Professor: Adam Oberman
Class: Monday, Wednesday, 10:05 am-11:25 am BURN 1214
Office Hours Fall 2022: BURN 1106, MW 11:30-12:00, and by appointment. Additional office hours TBA.
Math and Stats Majors/Honours students, Computer Science students.
A mathematically rigorous approach to Machine Learning (ML). This course will cover the mathematical models which go into current machine learning models, as well as deep learning architectures, and application areas. It will provide the necessary background for understanding deep learning models and reading contemporary research papers. The sequel, Math 562, will go more deeply into mathematical aspects, such as statistical learning theory, and regularization.
08/31 (Weds): Discuss course, AI vs ML, Clustering intro
09/07 (Weds): k-means clustering, losses, hypothesis classes
09/12 (Monday): Vector Calculus Review, Vector Calc for ML: sigmoids. 09/14 (Wednesday): Vector Calc for ML. Generative and discriminative models.
Homework 1, posted Sept 20, due Sept 27 (in myCourses) (updated Sept 27)
09/14 (Monday) and 09/21 (Weds)
09/28 (Wednesday): Programming languages for the class, using Python and Collab. K-means code.
Homework 2, revised Oct 7
10/13 (Thursday) first lecture after reading break
Homework 3, final version posted.
10/24 (Monday), 10/26 (no lecture), 10/31 (Monday)
11/7 (Monday): Features (handwritten notes)
11/9 (Weds): midterm
RKHS Ch 12 of M. J. Wainwright (2019) High-dimensional statistics: A non-asymptotic viewpoint. RKHS Ch 12 of M. J. Wainwright (2019) High-dimensional statistics: A non-asymptotic viewpoint.
We covered 12.1 Hilbert Space Basics, and 12.2.1 PSD kernel functions
HW 4 Posted
HW 5 HW5.pdf