18.408: Algorithmic Aspects of Machine Learning

Spring 2023



Modern machine learning systems are often built on top of algorithms that do not have provable guarantees, and it is the subject of debate when and why they work. In this class, we will focus on designing algorithms whose performance we can rigorously analyze for fundamental machine learning problems. We will cover topics such as: nonnegative matrix factorization, tensor decomposition, learning mixture models, graphical models, community detection, matrix completion and robust statistics. Almost all of these problems are computationally hard in the worst-case and so developing an algorithmic theory is about (1) choosing the right models in which to study these problems and (2) developing the appropriate mathematical tools (often from probability, geometry or algebra) in order to rigorously analyze existing heuristics, or to design fundamentally new algorithms.

Announcement: Project guidelines are posted here and due May 16th, after class.

Announcement: PSET 2 is posted here and due April 6th, after class.

Announcement: PSET 1 is posted here and due March 7th, after class.

Course Information

Instructor Notes

Course Outline

Here is a tentative outline for the course: