Description:

Algorithms for Nonconvex, Nonsmooth Optimization

Featured Speaker:
Frank E. Curtis
Asscoaite Professor, Industrail and Systems Engineering
Lehigh University

Abstract: Intuitive descriptions of algorithms for solving nonconvex, nonsmooth optimization problems are presented. Starting with fund- amentals of subgradient, cutting plane, and gradient sampling methods for solving convex problems, the groundwork is laid for advancing to modern adaptive methods for solving more challenging nonconvex problems. Algorithms are presented from a geometric viewpoint. The tutorial closes with a discussion of issues related to the implementation of adaptive methods in the NonOpt software package, currently under development in collaboration with Andreas Wächter (Northwestern University)