Scenario optimization for probabilistically robust design: foundations and recent developments.
Event details
| Date | 13.05.2016 |
| Hour | 10:15 › 11:15 |
| Speaker | Prof. Giuseppe Calafiore, Politecnico di Torino |
| Location | |
| Category | Conferences - Seminars |
Scenario optimization, a.k.a. random convex programming, gained increasing recognition among researchers and practicioners interested in design under uncertainty (in various fields, ranging from finance to robust control), due to its computational efficiency, flanked by strong theoretical properties. In this seminar, we shall introduce scenario design as a design technique that sits in between worst-case deterministic design and chance-constrained design.
We shall describe a fundamental result that ties together three key quantities:
(i) the complexity of the model to be optimized,
(ii) the number of training samples of the uncertainty (scenarios) used in the design, and
(iii) the probabilistic robustness of the ensuing model.
Discussion and application examples on this seminal result will lead us to introduce and describe a novel paradigm, called “repetitive scenario design,” which generalizes the classical scenario method and allows for wider practical applicability.
Bio: I received the “Laurea” degree in Electrical Engineering from Politecnico di Torino in 1993, and the Ph.D in Information and System Theory from Politecnico di Torino, in 1997. Since 1998 I am with the faculty of Dipartimento di Automatica e Informatica, Politecnico di Torino, where I am currently serving as a Full Professor.
I held several visiting positions at international institutions: at Information Systems Laboratory, Stanford University, California, in 1995; at Ecole Nationale Supérieure de Techniques Avanceés (ENSTA), Paris, in 1998; and at the University of California at Berkeley, in 1999, 2003 and 2007, at the Institute of Pure and Applied Mathematics, University of California at Los Angeles, 2010, and at the John Von Neumann Institute, Vietnam National University in 2011. I was Associate Editor for the IEEE Transactions on Systems, Man, and Cybernetics (T-SMC), for the IEEE Transactions on Automation Science and Engineering (T-ASE), and for Journal Européen des Systèmes Automatisés.
I am the author of more than 150 journal and conference proceedings papers, and of 8 books.I am the coordinator of a Specializing Master in Automatica and Control Technologies (MACT) at Politecnico di Torino. My research interests are in the fields of convex optimization, randomized algorithms, identification, and control of uncertain systems, with applications ranging from finance and economic systems to robust control, pattern recognition and robotics.
We shall describe a fundamental result that ties together three key quantities:
(i) the complexity of the model to be optimized,
(ii) the number of training samples of the uncertainty (scenarios) used in the design, and
(iii) the probabilistic robustness of the ensuing model.
Discussion and application examples on this seminal result will lead us to introduce and describe a novel paradigm, called “repetitive scenario design,” which generalizes the classical scenario method and allows for wider practical applicability.
Bio: I received the “Laurea” degree in Electrical Engineering from Politecnico di Torino in 1993, and the Ph.D in Information and System Theory from Politecnico di Torino, in 1997. Since 1998 I am with the faculty of Dipartimento di Automatica e Informatica, Politecnico di Torino, where I am currently serving as a Full Professor.
I held several visiting positions at international institutions: at Information Systems Laboratory, Stanford University, California, in 1995; at Ecole Nationale Supérieure de Techniques Avanceés (ENSTA), Paris, in 1998; and at the University of California at Berkeley, in 1999, 2003 and 2007, at the Institute of Pure and Applied Mathematics, University of California at Los Angeles, 2010, and at the John Von Neumann Institute, Vietnam National University in 2011. I was Associate Editor for the IEEE Transactions on Systems, Man, and Cybernetics (T-SMC), for the IEEE Transactions on Automation Science and Engineering (T-ASE), and for Journal Européen des Systèmes Automatisés.
I am the author of more than 150 journal and conference proceedings papers, and of 8 books.I am the coordinator of a Specializing Master in Automatica and Control Technologies (MACT) at Politecnico di Torino. My research interests are in the fields of convex optimization, randomized algorithms, identification, and control of uncertain systems, with applications ranging from finance and economic systems to robust control, pattern recognition and robotics.
Links
Practical information
- Expert
- Free