70. Smoothing Effect of Epi-convergence and Inf-convolution in Optimization
Invited abstract in session TD-3: Variational techniques and subdifferentials, stream Variational analysis: theory and algorithms.
Thursday, 14:10 - 15:50Room: M:J
Authors (first author is the speaker)
| 1. | Mohammed Ibrahim Abdelkayoum Ghitri
|
| Mathematics, Universidad de Alicante | |
| 2. | Abderrahim Hantoute
|
| Mathematics, Universidad de Alicante |
Abstract
We use smoothing processes based on the infimal convolution of convex, proper and lower semi-continuous functions to regularise optimization problems given by convex-composite functions. We show that the proposed regularisation schemes still (epi-)converge to the original data, even if the chosen kernel is an arbitrary convex function. This also allows to derive upper estimates of the subdifferentials of the epi-limits of non-convex functions without using qualifying conditions (namely BCQ-type conditions).
Keywords
- Convex and non-smooth optimization
Status: accepted
Back to the list of papers