Home » Node » 19981

Derivative-free methods for black-box optimization problems (public)

speaker DIAG: 
Data dell'evento: 
Saturday, 23 May, 2020 - 10:00
Luogo: 
Online paitaforma Meet codice accesso woh-ufpo-zav
Contatto: 
laura.palagi@uniroma1.it

Giampaolo LIUZZI è risultato vincitore della procedura selettiva per n.1 posto di ricercatore a tempo determinato – tipologia B ai sensi dell’art.24, comma 3, lett. b, legge 240/2010 – per il settore concorsuale 01/A6 - settore scientifico disciplinare MAT/09 - codice concorso 2019RTDB006, bandito con Decreto Rettorale D.R. n. 1905/2019 del 21.06.2019, i cui atti sono stati approvati con Decreto Rettorale D.R. n. 1351/2020 del 19/05/2020.

Nell'ambito della procedura ai fini della chiamata da parte del Consiglio di dipartimento, Giampaolo LIUZZI terrà un seminario pubblico sulle attività di ricerca da lui svolte e in corso di svolgimento. Il seminario sarà svolto in modalità telematica sabato 23 maggio 2020 alle ore 10:00. Per partecipare, connettersi all’indirizzo meet.google.com/woh-ufpo-zav

Abstract

In this talk, I will present my recent and past research activity which has been mainly focused on the definition and study of methods for the solution of nonlinear optimization problems. The main results obtained for nonlinearly constrained optimization problems will be briefly revised. I will describe more in detail the research activity carried out in the field of derivative-free methods for black-box problems. Such problems are ubiquitous in many fields. They arise, for instance, when one wants to optimize the performances of a complex physical system through the use of simulation programs. In such a context, i.e. when derivatives of the objective and/or constraint functions can be neither calculated nor approximated explicitly, further difficulties can arise which must be taken into proper account to come up with algorithms which are both theoretically sound and numerically efficient. I will present some methods which, when possible, take advantage of the particular structure of the problem. Most of these state-of-the-art methods have been made available in an open-access Derivative-Free Library.

Short bio

Giampaolo Liuzzi received his M.Sc. in Engineering in Computer Science in 1997 from Sapienza University of Rome. From 1998 to 2000 he followed a Ph.D. program in Operations Research at the Department of Computer, Control, and Management Engineering "Antonio Ruberti" at Sapienza University of Rome, and received his Ph.D. degree in 2001 with a dissertation new methods for constrained nonlinear programming. Since 2010, he has a permanent position as a researcher at the Institute for Systems Analysis and Computer Science “Antonio Ruberti” (IASI) of the National Research Council (CNR). He is also involved with teaching activity mainly for the Department of Computer, Control, and Management Engineering, since 2000. His research interests include nonlinear programming, derivative-free methods, multiobjective optimization, and global optimization. He published 48 papers in international journals and he has also been involved in practical problems like, e.g., the optimal design of magnetic resonance equipments, the optimization of electric motors and integrated circuits, the optimal design of ship hulls.

© Università degli Studi di Roma "La Sapienza" - Piazzale Aldo Moro 5, 00185 Roma