Department of Information and Computing Sciences

Departement Informatica Onderwijs
Bachelor Informatica Informatiekunde Kunstmatige intelligentie Master Computing Science Game&Media Technology Artifical Intelligence Human Computer Interaction Business Informatics

Onderwijs Informatica en Informatiekunde

Vak-informatie Informatica en Informatiekunde

Multi-agent systems

Website:website containing additional information
Course code:INFOMAS
Credits:7.5 ECTS
Period:period 3 (week 6 through 15, i.e., 8-2-2021 through 16-4-2021; retake week 27)
Participants:up till now 59 subscriptions
Schedule:Official schedule representation can be found in MyTimetable
lecture          Mehdi Dastani
Tim Baarslag
Contents:This course focuses on multiagent issues and will consist of lecture and tutorial sessions.

  • The lectures will cover the following topics:
    • Game theory
    • Communication
    • Social choice
    • Mechanism Design
    • Auctions
  • The tutorial sessions are organised around a student group assignment to design a negotiating agent. The group assignment consists of three written reports and a Java implementation of the negotiation agents. The assignment covers multiagent system issues such as:
    • Multi-agent negotiation
    • Preference modeling and utility theory
    • Communication protocols and coordination in Multi-Agent Systems
    • Group decision-making
    • Opponent modeling
    • Decision-making under uncertainty
    • Development of Multi-Agent Systems
  • Literature:We use parts of the following books and some additional papers. The first book is available on the internet, see: The second book is available in the library and in my own room (BBG-521).
    Course form:Lectures + Tutorial
    Exam form:Written Exams + Assignment
    Minimum effort to qualify for 2nd chance exam:Om aan de aanvullende toets te mogen meedoen moet de oorspronkelijke uitslag minstens 4 zijn.
    Description:Multiagent systems are nowadays an important and innovative research area in artificial intelligence and computer science. They are viewed as a computational paradigm to specify, design, and implement software systems for solving complex tasks in terms of interacting computational entities. These computational entities, called agents, are usually assumed to be heterogeneous, autonomous, self-interested, and situated in an environment. They can communicate and coordinate their activities to achieve their objectives. The environments in which individual agents are situated can be ordinary software environments (e.g., internet), physical environments (e.g., robot), or normative mechanisms (e.g., virtual organisation). Multiagent system paradigm can be used to develop open and distributed applications such as multi-robot systems, on-line auctions, virtual organizations, electronic institutions, market places, and virtual games. Some of multiagent issues are: how agents communicate and interact with each other, how they coordinate their activities, how their activities can be organised, and what are characteristics of multiagent applications.