Carnegie Mellon University

Modern Convex Optimization

Course Number: 47851

The purpose of this course is to give a solid foundation on convex optimization. Convex optimization, broadly speaking, is the most general class of optimization problems that are efficiently solvable. It has been fundamental in the development of Operations Research based decision making, and it naturally arises and is successfully used in a diverse set of applications in machine learning and high-dimensional statistics, signal processing, control, engineering, medical imaging, etc. In this course, we will focus on the modern aspects of convex optimization beyond Linear Programming, such as conic optimization including quadratic programming and semidefinite programming, and the main templates for efficient algorithms to solve large-scale convex optimization problems. It will concentrate on recognizing and solving convex optimization problems with an in-depth study of the underlying theory. The aim is to give students the background required to use these methods in their own research, and therefore our focus will be on the recent developments.

Degree: PhD
Concentration: Operations Research
Academic Year: 2019-2020
Semester(s): Mini 3
Required/Elective: Elective
Units: 6


Lecture: 100min/wk and Recitation: 50min/wk