Discover the surprising connection between the convex-concave procedure (CCCP) and the Frank-Wolfe (FW) method in this 24-minute conference talk from the Workshop on "One World Optimization Seminar in Vienna" at the Erwin Schrödinger International Institute for Mathematics and Physics. Explore how CCCP and its generalization to constrained problems are special cases of the Frank-Wolfe method, providing deep pedagogical insights and transferring nonconvex Frank-Wolfe convergence theory to CCCP. Learn how this perspective may facilitate the application of other FW advances to CCCP and its generalizations, potentially advancing the field of optimization.
CCCP is Frank-Wolfe in Disguise - Connecting Optimization Methods
Erwin Schrödinger International Institute for Mathematics and Physics (ESI) via YouTube
Overview
Syllabus
Alp Yurtsever - CCCP is Frank-Wolfe in disguise
Taught by
Erwin Schrödinger International Institute for Mathematics and Physics (ESI)
Related Courses
-
Proximal Gradient Methods for Nonsmooth Nonconvex Minimax - A Unified Convergence Analysis
-
Nonnegative Polynomials, Nonconvex Polynomial Optimization, and Applications to Learning
-
Nonconvex Optimization and the Curse of Local Minima - Lessons Learned from Orbital Tomography
-
Exploiting Geometric Structure in Matrix-Valued Optimization
-
Nonmonotone Forward-Backward Splitting Method for Infinite-Dimensional Optimization Problems
-
Stochastic Primal Dual Splitting Algorithms for Convex and Nonconvex Composite Optimization in Imaging
