The objective of this course is to give graduate students an introduction to nonlinear programming (specifically focusing on convex optimization). Students are expected to have proper knowledge in linear algebra, calculus, linear programming, and other basic OR techniques which are normally expected to be covered in the undergraduate operations research curriculum.

MATH110 Calculus, MATH112 Applied Linear Algebra, IMEN261 Introduction to Operations Research or equivalent

Attendance 10%
Homework 0%
Algorithm Coding 10%
Midterm 40%
Final 40%
TOTAL 100%

Week 1-8 Chapter 1-5: Theory on Convex Optimization

Week 9-11 Chapter 9: Unconstrained minimization

Week 12-13 Chapter 10: Equality constrained minimization

Week 14-16 Chapter 11: Interior-point methods

Roughly, homework assignments will be assigned when each chapter finishes. They are not collected. The problems in them are for students’ practice, and some of them may appear in the midterm and the final exams.

There will be two exams for this course. Exams may or may not be cumulative. A university excused absence is required to make up a missed exam and arrangements should be made prior to the testing day. Other instructions regarding exams will be given in class.

There will be two algorithm coding assignments implementing the gradient and Newton methods. Students will practice implementing two algorithms from scratch using Python or Julia language, which will help them understand those algorithms.