Course Description

An introductory course in machine learning, with a focus on data modeling and related methods and learning algorithms for data sciences. Tentative topic list:

• Dimensionality reduction, such as principal component analysis (PCA) and the singular value decomposition (SVD), canonical correlation analysis (CCA), independent component analysis (ICA), compressed sensing, random projection, the information bottleneck. (We expect to cover some, but probably not all, of these topics).
• Clustering, such as k-means, Gaussian mixture models, the expectation-maximization (EM) algorithm, link-based clustering. (We do not expect to cover hierarchical or spectral clustering.).
• Probabilistic-modeling topics such as graphical models, latent-variable models, inference (e.g., belief propagation), parameter learning.
• Regression will be covered if time permits.

Can be taken independently or in any order with CS4780/5780 (Machine Learning for Intelligent Systems).

Prerequisites: probability theory (BTRY 3080, ECON 3130, MATH 4710, or strong performance in ENGRD 2700 or equivalent); linear algebra (MATH 2940 or equivalent); CS2110 or equivalent programming proficiency.

News (see also announcements on lecture handouts)

• Competition 1 has begun. Due 4th may. Competition 1
• Monday, February 22: Assignment 1 is out and is due on march 7th 11:59pm. Start forming groups soon.
• Thursday, January 28: clarification regarding CMS: we won't be adding any students to it until after HW0 is due, and then we'll do a big semi-auto-enroll from the registrar's records and what gets handed in. So, no need to ask us to add you to CMS.
• Thursday, January 28: The first question has been asked and answered on the course Piazza page. We encourage you to use this resource to ask and answer questions.
• Thursday, January 28: The first assignment is out!