Honors Undergraduate Course on Machine Learning
Math 3094, Spring Semester 2021
University of Connecticut
The interdisciplinary field known as Machine Learning or Data Science draws together techniques from computer science, mathematics, and statistics to extract meaning from data. In this course, we will discuss some of the essential mathematical ideas in this field.
While our focus will be on the role of Calculus, Probability, and Linear Algebra, we will introduce computational techniques using Python and the Jupyter notebook environment, and some ideas from statistics, in order to closely link theory and practice.
The course will meet synchronously online on Tuesdays and Thursdays from 11:00 to 12:15 EST using BlackBoard Collaborate through the HuskyCT UConn LMS.
An outline of the course topics is available here.
A guide to the first assigned project is here
We will use the Campus Wire platform for online help and discussions. Students enrolled in the course should receive an electronic invite to the forum. Contact one of the professors if you need access.
We will rely on the Python programming language, the Anaconda open source data science platform, and the Jupyter notebook environment for our computer work. All of this software can be obtained for Linux, Mac, or Windows from the Anaconda website: www.anaconda.com.
A very brief guide to installing the software is available here.
There is no official textbook for the course. We will be providing notes as we progress. The following texts may be useful as references.
James, Witten, Hastie, Tibshirani. An Introduction to Statistical Learning (with Applications in R). This is an introductory text on machine learning with a more statistical emphasis than our course, and with computer examples in R instead of Python. It is an excellent and informative work, and it is available for free from the book home page.
Bass, Alonso-Ruiz, Baudoin, et. al.
UConn’s Open Undergraduate Probability Text. This is the (open source) textbook for UConn’s undergraduate probability course, Math 3160.
Boyd, S. and Vandenberghe, L. Introduction to Applied Linear Algebra. This is a (free) introductory text on Linear Algebra with a focus on applications, especially to Least Squares.
Treil, S. Linear Algebra Done Wrong. This is a more theoretical linear algebra text that treats important topics such as inner product spaces.
Bishop, C. Pattern Recognition and Machine Learning This is a (free) comprehensive look at machine learning; it claims to be aimed at “advanced undergraduates or first year PhD students” but is technically demanding.