Seoul National University
4190.773 Optimization for Machine Learning
(Special Topics in Artificial Intelligence)
Spring 2018 - U Kang

News and Announcements

Course Information

Optimization is a crucial tool for many machine learning techniques. Formulating a problem into an optimization framework, and solving it are core skills for researchers in the area of machine learning. This course covers important theories and algorithms for optimization in machine learning. Topics include convex sets, convex functions, convex optimization, duality, submodular optimization, and algorithms for optimizations.

Schedule

Date Topic Notice
Mar 5 Introduction
7 Preliminaries: Linear Algebra
12 Preliminaries: Math
14 Preliminaries: Math-2: use the previous slide
19 Introduction to Convex Optimization
21 Convex Sets
26 Convex Sets-2
28 Convex Sets-3 HW 1 out
Apr 2 Convex Functions
4 Convex Functions-2
9 Convex Functions-3
11 Convex Functions-4 HW 2 out
16, 18 Midterm
23 Convex Optimization Problems
25 Convex Optimization Problems-2
30 Convex Optimization Problems-3
May 2 Convex Optimization Problems-4
7 Convex Optimization Problems-5 HW 3 out
9 Duality
14 Duality-2
16 Duality-3
21 Duality-4 HW 4 out
23 Submodularity
28 Submodularity-2
30 Submodularity-3
June 4 Algorithms for Optimization
6 Algorithms for Optimization-2
11 Conclusion
13 Final

Grading

Late policy - for all deliverables:

Textbook

The text book is
Convex Optimization by Stephen Boyd and Lieven Vandenberghe. (available in online)

Prerequisite

We expect you to have an undergraduate-level knowledge on the following topics: We provide some background, but the class will be fast paced.
Last modified Mar. 1, 2018, by U Kang