64. Exploring Polynomial Models in the Search Step of Direct Multisearch
Invited abstract in session WC-1: Advances in Multiobjective and Bilevel Optimization without Derivatives, stream Zeroth and first-order optimization methods.
Wednesday, 14:00-16:00Room: B100/1001
Authors (first author is the speaker)
| 1. | Marta Pozzi
|
| University Bocconi, University Bocconi | |
| 2. | Everton Silva
|
| Mathematics, NOVA School of Science and Technology | |
| 3. | Ana Luisa Custodio
|
| Dept. Mathematics, Universidade Nova de Lisboa |
Abstract
Direct Multisearch (DMS) is a direct search method suited for multiobjective derivative-free optimization. Its structure is organized in an optional search and a poll step, the latter responsible for convergence.
Quadratic models were recently introduced in the definition of the search step, improving the numerical performance of the algorithm. In this work, new strategies that make use of these models are explored and evaluated.
The MinMax scalarization approach is compared with the Epsilon and NBI methods, and with a strategy that uses the recently released Improved Steepest Descent algorithm.
Keywords
- Derivative-free optimization
- Multi-objective optimization
Status: accepted
Back to the list of papers