Numerical Optimization

Catalog description

Many problems in science, engineering, medicine and business involve optimization. in which we seek to optimize a mathematical measure of goodness subject to constraints. This course will cover the basics of smooth unconstrained and constrained optimization in one and more variables: first and second order conditions, Lagrange multipliers, KKT conditions, Gradient descent, Newton and Quasi-Newton methods. .Key concepts and methods in mathematical programming will then be covered: linear programming, quadratic and convex programming (simplex method, primal-dual methods, interior point methods) with applications to engineering, optimal control and machine learning. Prerequisite Course(s): Prerequisites: MATH 410 Classroom Hours - Laboratory and/or Studio Hours – Course Credits: 3-0-3