Teaching‎ > ‎


AMATH 515, Optimization: Fundamentals and Applications

University of Washington, Winter 2016.
  • Pre-requisites: Linear Algebra and Advanced Calculus/Analysis. 
  • Recommended: Probability and Statistics. Scientific Programming in a language such as Matlab or Julia is desirable.
  • We meet MWF 9:30 - 10:20, SIG 230.

Quick Links


The course will cover fundamental concepts in optimization, with focus on applications. We will start with basic convex analysis, conjugacy, and Fenchel duality. We will then proceed to study a range of applications, discussing modeling, algorithms, and problem structure. In particular, we will look at sparse and robust regression, nonlinear inverse problems, time series applications, and topics in machine learning (including neural nets).

Announcements and Discussions

Teaching Assistant

  • Peng Zheng (zhengp@uw.edu), Office Hours:


There will be 4 assignments throughout the term. All homework assignments must be written in LaTex. The grade for each assignment will be a combination of points for completeness, and points for graded problems. You are welcome to discuss and work on homework problems together. However, please write up the solutions on your own. 

Reference Materials

We will make use of materials available online.  The list of materials will grow as we proceed. 


Machine learning: 


We will use Matlab and Matlab packages for class and homework. 
  • If you do not have Matlab and do not want to purchase it, you can use Octave or other computational languages (e.g. Julia). 
  • You should be aware of CVX, a Matlab package that allows you to solve a wide variety of convex optimization problems. 
  • We may also use minFunc, a package written by Mark Schmidt. 

Final Project

There will be a final (group) project, with details to be announced on this website. The point of the group project is to have you go in depth into a topic of your choice. The project includes a written component, a coding component, and a presentation to the group. 


Your grade will be determined by homework, participation in class, and a final project. Participation in class includes attendance, as well as completing exercises in lecture. 

Homework assignments:               60% 
Class participation:                          10%
Final project:                                      30%

Academic Honesty Policy

Please read the policy here. By staying registered in the class you indicate your acceptance of all its terms. We do not accept late homework or absence without official reasons (medical, etc.) approved by a student dean. If you miss class, please coordinate with colleagues to find out what you missed (do not email the professor to help you catch up). 

Subpages (2): homework lectures
Sasha Aravkin,
Jan 4, 2016, 6:45 AM