Learning in Collaborative Multi-Agent Systems

Learning in Collaborative Multi-Agent Systems

Lecturer: Dr.-Ing. Nugroho Fredivianus

Hours per week: 2 for lecture + 2 for exercise

Language: English

Schedules:

Lectures: every Tuesday 14:00 - 16:00, Room 1332, Start: 25.10.2016
Exercises: every Thursday 08:00 - 10:00, Room -1319, Start: 27.10.2016

Examination: oral exam (25 min.) or written exam (120 min.)

Schedule:tbd
Room: tbd

Assigned for: Informatik, Elektrotechnik, Wirtschafts-Ing., Mechatronik

Course of study: Master (M.Sc.)

Number of participants: max 20 students

Goal: The understanding of collaborative distributed systems esp. Multi-Agent Systems (MAS), whose intelligence obtained after performing a specific Machine Learning method, e.g., decentralized market control, a team of robotic soccer.

Content:

  • Agent model and Self-X properties
  • Collaboration and competition in Multi-Agent Systems
  • Nature-inspired algorithms
  • Machine learning, esp. reinforcement learning
  • Real application examples: robotic soccer team and more

Additional information: the exercises are mainly based on programming in NetLogo (free Multi-Agent Simulator), downloadable at http://ccl.northwestern.edu/netlogo/index.shtml

Literature: the following literature will be extended during the lectures:

  • Gerhard Weiss (ed.): Multiagent Systems – A Modern Approach to Distributed Artificial Intelligence
  • Mohri Mehryar et al: Foundations of Machine Learning
  • Eric Bonabeau et al: Swarm Intelligence – From Natural to Artificial Systems
  • Sven Brueckner et al (ed.): Engineering Self-Organising Systems: Methodologies and Applications
  • Weiming Shen et al: Multi-Agent Systems for concurrent Intelligent Design and Manufacturing

Attention: all lecture slides will be available at the Moodle site.
The link to the course/lecture in HIS-LSF can be find here.