The Scenario Approach for Decision and Control

2024 CDC full-day Workshop
Speakers: Marco C. Campi, University of Brescia (organizer)
Algo Carè, University of Brescia (organizer)
Simone Garatti, Politecnico di Milano (organizer)
Kostas Margellos, University of Oxford
Maria Prandini, Politecnico di Milano
Date: Sunday, December 15, 2024
Time: 8:45-17:30
Location:  to be announced

Workshop Structure

Schedule (full-day workshop)

08.45-09.00:
Introduction and motivation
09.00-10.30:
Simone Garatti: The Scenario Approach in the dawn of sample-based and data-driven decision-making
10.30-11.00:
Coffee break
11.00-12.00:
Maria Prandini: Optimal constrained control of stochastic linear systems operating in stationary conditions
12.00-13.30:
Lunch break
13.30-14.30:
Marco C. Campi: Scenario Optimization -- The boon of direct design and the marriage between risk and complexity for reliability certification
14.30-15.30:
Algo Carè: Guaranteed supervised classification for medical applications
15.30-16.00:
Coffee break
16.00-17.30:
Kostas Margellos: Data privacy and outlier removal in multi-agent scenario optimization


Workshop Presentation

This one-day workshop introduces participants to the scenario approach, a well-established methodology for data-driven decision-making with applications in control design, machine learning, and prediction.

Over the past two decades, the scenario approach has evolved from determining sample sizes in randomized control design to a comprehensive framework for utilizing data in various engineering fields. The workshop brings together leading experts who have significantly contributed to these developments.

Participants will gain access to state-of-the-art advancements in scenario theory, addressing critical challenges in modern control design and decision-making. In particular, the workshop will provide theory and algorithms for:

   i. establishing a-priori guarantees and finite sample complexity bounds for optimization based decision-making;
   ii. exploring the trade-off between (probabilistic) feasibility and performance;
   iii. deriving a-posteriori guarantees for decision making problems including non-convex programs;
   iv. establishing connections to statistical learning theory; and
   v. addressing applications to multi-agent optimization with agent-private data-sets.

The workshop aims to foster innovation by exploring new directions and challenges in the field.

Ideal for graduate students, advanced control engineers, and researchers interested in systems and control, machine learning, and decision-making, the workshop covers both theoretical and practical perspectives.

This workshop promises an insightful and productive day of learning and exchange.  Lively interactions between our speakers and the participants will be encouraged. 




For more information, please write to: algo.care@unibs.it