High-Dimensional Probability and Statistics

MATH/STAT/ECE 888 - Topics in Mathematical Data Science (Fall ’21)
Sebastien Roch, Department of Mathematics, UW-Madison

In Fall 2021, this course will provide a rigorous, self-contained introduction to the area of high-dimensional probability and statistics from a non-asymptotic perspective, aimed at graduate students in mathematics, statistics, computer science and engineering. It will include a focus on core methodology and theory (tail bounds, concentration of measure, random matrices) as well as in-depth exploration of particular model classes (sparse linear models, graphical models, community detection). No statistics background will be assumed. Prior exposure to graduate-level probability (e.g., MATH 733 or ECE 730 or STAT 709) is highly recommended.

Course Information

Course: MATH/STAT/ECE 888 - Topics in Mathematical Data Science
Semester: Fall 2021
Time: MoWeFr 2:25PM - 3:15PM
Place: VAN VLECK B239
Instructor: Sebastien Roch
Email: Use Canvas for all communications

Textbooks

Topics to be covered will be taken partly from the following textbooks (but see also the full list of references):

Vershynin, R. (2018). High-Dimensional Probability: An Introduction with Applications in Data Science. CUP.
Wainwright, M. J. (2019). High-Dimensional Statistics: A Non-Asymptotic Viewpoint. CUP.

Syllabus

A full syllabus is here.

Grades will be based on:

Scribing one lecture. Details are here.
One or two homework assignments. They will be posted on Canvas.
Writing a short report on a workshop talk of relevance. A list of virtual workshops will be posted on Canvas.

COVID Policies

Students enrolled in this class are expected to comply with the University’s current COVID rules that are maintained here: https://covidresponse.wisc.edu