501. A multilevel stochastic regularized first-order method with application to training
Invited abstract in session TA-35: Nonlinear Optimization Algorithms and Applications: 4, stream Continuous and mixed-integer nonlinear programming: theory and algorithms.
Tuesday, 8:30-10:00Room: Michael Sadler LG15
Authors (first author is the speaker)
| 1. | Margherita Porcelli
|
| Dipartimento di Ingegneria Industriale, Università degli Studi di Firenze | |
| 2. | Filippo Marini
|
| DIEF - Dipartimento di Ingegneria Industriale, Università degli Studi di Firenze | |
| 3. | Elisa Riccietti
|
| LIP, ENS Lyon |
Abstract
We present a new multilevel stochastic framework for the solution of optimization problems. The proposed approach uses random regularized first-order models that exploit an available hierarchical description of the problem, being either in the classical variable space or in the function space, meaning that different levels of accuracy for the objective function are available. We present the converge analysis of the method and show its numerical behavior on the solution of finite-sum minimization problems arising in binary classification problems.
Keywords
- Continuous Optimization
- Stochastic Optimization
Status: accepted
Back to the list of papers