Chaque module vaut 3 ECTS. Vous sélectionnez 10 modules/30 ECTS parmi les catégories suivantes:
- 12-15 crédits ECTS en Modules technico-scientifiques (TSM)
Les modules TSM vous transmettent une compétence technique spécifique à votre orientation et complètent les modules de spécialisation décentralisés. - 9-12 crédits ECTS en Bases théoriques élargies (FTP)
Les modules FTP traitent de bases théoriques telles que les mathématiques élevées, la physique, la théorie de l’information, la chimie, etc., vous permettant d’étendre votre profondeur scientifique abstraite et de contribuer à créer le lien important entre l’abstraction et l’application dans le domaine de l’innovation. - 6-9 crédits ECTS en Modules contextuels (CM)
Les modules CM vous transmettent des compétences supplémentaires dans des domaines tels que la gestion des technologies, la gestion d’entreprise, la communication, la gestion de projets, le droit des brevets et des contrats, etc.
Le descriptif de module (download pdf) contient le détail des langues pour chaque module selon les catégories suivantes:
- leçons
- documentation
- examen
Natural, social, and engineered complex systems can be modelled as being composed of agents interacting with one another and their environment. This course introduces students to the theory, tools and techniques for understanding and solving problems related to such systems.
The course is composed of two parts. In the first one, both cooperative and selfish agents and interactions between them will be discussed. The methodological support will be provided by game theory.
In the second part, the focus will be on the study and analysis of models of systems in the aim of understanding the conditions under which certain properties can emerge and agent might learn certain strategies or behaviours by interacting with the environment and themselves.
Throughout the course, several application areas such as cooperation and competition, social influence and reinforcement learning will be discussed.
Compétences préalables
Basic knowledge of probability, algebra, calculus and differential equations. Basics of procedural programming and ability to implement small programs in an arbitrary language, e.g. Python, Matlab, R, Java, C#, C++, C, etc.
Objectifs d'apprentissage
A successful participant of this course is able to
- understand the rationale of multi-agent systems and their modelling.
- model scenarios with multiple interacting agents in the language of game theory
- evaluate the feasibility of achieving goals with agents using game theory
- understand the basic approaches to multi-agent learning, their peculiarities and their differences
- learn to choose the appropriate class of models with agents to characterise different complex systems
- implement in an efficient way a model of a system, then understand and analyse the corresponding outputs
Contenu des modules
- Review of single-agent decision making and learning
- Multi-agent interaction:
- games in normal form, dominant strategies,
- Nash equilibria, Pareto optimality,
- partial observability,
- cooperative and coalition games, Shapley value,
- repeated games,
- Multi-agent learning:
- model based approaches: fictitious and rational learning
- model-free approaches: no regret and reinforcement learning
Méthodes d'enseignement et d'apprentissage
- Lectures
- Exercises and homework
- Practical work with appropriate tools
- Literature studies
Bibliographie
- A Concise Introduction to Multi-Agent Systems and Distributed Artificial Intelligence. Nikos Vlassis. Morgan & Claypool Publishers, 2007.
- Introduction to Multi-Agent Systems - 2nd Edition. Michael Wooldridge. John Wiley & Sons, 2009.
- Multi-Agent Systems. Yoav Shoham and Kevin Leyton-Brown. Cambridge University Press, 2009.
- Artificial Intelligence, A Modern Approach (4th Edition). Stuart Russell and Peter Norvig. Pearson. 2021
Télécharger le descriptif complet
Retour