|
Academic Year |
2023/2024 |
Name |
Dott. - MI (1380) Ingegneria dell'Informazione / Information Technology |
Programme Year |
1 |
ID Code |
061649 |
Course Title |
DISTRIBUTED ALGORITHMS FOR OPTIMIZATION AND CONTROL OVER NETWORKS |
Course Type |
MONO-DISCIPLINARY COURSE |
Credits (CFU / ECTS) |
5.0 |
Course Description |
Energy systems and transportation systems are examples of large-scale systems involving multiple agents aiming at taking decisions in the most profitable way, while interacting with each other.
Motivated by such applications, this course will present a mathematical framework for modeling, algorithm design and analysis for cooperative decision making problems arising in multi-agent systems. We will start from basic concepts in optimization, and will then apply them to distributed multi-agent dynamics seeking convergence to cooperative equilibria.
1 - Motivation and illustrative applications
Description of decision making problems arising in smart grid control and optimization, and in the coordination and control of electric vehicle fleets. These applications will set a common ground to highlight the main complexity issues arising in multi-agent systems, namely, scalability due to the large scale nature of these problems, heterogeneity of the agents, privacy, and uncertainty.
2 - Mathematical tools
Introduction to the mathematical tools that constitute the theoretical backbone for the analysis and design of cooperative algorithms.
Topics covered are:
Constrained convex optimization: basics of convex analysis and optimality conditions
Duality theory: Weak and strong duality, optimality conditions
Constrained optimization algorithms: projected gradient and proximal algorithm
Basic notions on graph theory and application to distributed averaging
These tools constitute the theoretical backbone for the analysis and design of cooperative algorithms that arise in multi-agent decision making problems.
3 - Distributed optimization algorithms
Algorithms for multi-agent decision making are presented, addressing decision-coupled and constraint-coupled problems, based on either primal-based or dual-based algorithms.
Algorithms for decision-coupled problems: Topics covered are the Augmented Lagrangian Method (ALM), the Alternating Direction Method of Multipliers (ADMM), the distributed projected subgradient, and the distributed proximal algorithm.Algorithms for constraint-coupled problems: Topics covered are dual decomposition, the dual subgradient, the distributed dual subgradient, and the distributed ADMM algorithm.
4 - Distributed optimization in uncertain networks
The algorithmic solutions described in Part 3 will be extended to the case when the multi-agent optimization problem is affected by uncertainty, which is known locally to each agent through a private set of data.
The extension of the scenario approach to a distributed setting will be described, thus leading to a data-driven solution. |
Scientific-Disciplinary Sector (SSD)
|
SSD Code
|
SSD Description
|
CFU
|
ING-INF/04
|
SYSTEMS AND CONTROL ENGINEERING
|
5.0
|
|
Alphabetical group
|
Name
|
Teaching Assignment Details
|
From (included)
|
To (excluded)
|
A
|
ZZZZ
|
Prandini Maria, Falsone Alessandro, Garatti Simone, Margellos Konstantinos Nektarios
|
|
|