222. Subgradient Methods for Minimizing Paraconvex Functions
Invited abstract in session FD-5: Structured nonsmooth optimization, stream Nonsmooth and nonconvex optimization algorithms.
Friday, 14:10 - 15:50Room: M:N
Authors (first author is the speaker)
| 1. | Morteza Rahimi
|
| Mathematics, University of Antwerp | |
| 2. | Masoud Ahookhosh
|
| Department of Mathematics, University of Antwerp |
Abstract
The primary objective of this talk is to study the convergence analysis subgradient methods for paraconvex functions which satisfy an appropriate error bound property and grow sharply away from its solution set. We establish that a linear convergence of these subgradient methods in the paraconvex setting. In these methods, the parameters regarding paraconvex and error bound properties only appear to find a valid initial point. As such, the linear rate of convergence is independence of these constant values. Some preliminary numerical result validate or theoretical foundations.
Keywords
- Convex and non-smooth optimization
Status: accepted
Back to the list of papers