Skip over navigation

Course Offerings

Course Evaluation Results

Course Details

Spring 2017-2018
ELE 538B   Graded A-F, P/D/F, Audit

Special Topics in Information Sciences and Systems - Large-Scale Optimization for Data Science

Yuxin Chen

Intro to optimization methods for large-scale problems in the context of data science and machine learning applications. We (1) explore algorithms efficient for smooth and nonsmooth problems, including gradient methods, proximal methods, ADMM, quasi-Newton methods, as well as large-scale numerical linear algebra; (2) discuss the efficacy of these methods in concrete data science problems (e.g. low-rank models, dictionary learning, graph matching), under appropriate statistical models; and (3) introduce a global geometric analysis to characterize the nonconvex landscape of the empirical risks in several estimation/learning problems.

Sample reading list:
Sebastien Bubeck, Convex Optimization: Algorithms and Complexity
Dimiri Bertsekas, Nonlinear Programming (3rd edition)

Requirements/Grading:
Design Project - 50%
Problem set(s) - 40%
Other (See Instructor) - 10%

Other Requirements:
Not Open to Freshmen.

Website:  http://www.princeton.edu/~yc5/ele538_optimization/index.html

Schedule/Classroom assignment:

Class numberSectionTimeDaysRoomEnrollmentStatus
43136 L01 09:30:00 am - 10:50:00 am M W   Sherrerd Hall   101   Enrolled:31 Limit:50