This paper outlines a single module in Artificial Intelligence (AI) that is offered in the Department of Computer Science and Information Systems at a College of Higher Education, (which was awarded university status in August 2005). The module focuses on established principles underlying the discipline and integrates practical work to help reinforce an understanding of what AI is trying to do.

We describe three main phases of multi-agent system education during a ten year period and analyze the practical exercises used during these phases. Combined with a description of the student population, pedagogical principles, and the course aims we draw conclusions as to the suitability of each of the exercises.

For the last three years, a course on multi-agent systems has been taught at University of Bath and is, in the context of an EU-project, being evolved into a more generic resource for courses in several Latin American countries. In this paper we report on the context of our course, its evolution, the problems of covering the wide range of material, approaches to assessment and student feedback. We are now entering a second phase of course development in the framework of the @LIS-TechNET project, where we are developing e-learning materials for teaching MAS in conjunction with other universities in Latin America and Europe, upon which we include a preliminary report.

We present the ARES 2 system, a simulator of a city struck by an earthquake to be used as testbed for agents written as assignment of a basic multi-agent systems course. We briefly present the structure of this basic course, look at ARES 2 from the perspective of an instructor, a student having to implement agents running in it, and a system developer. We also report on our experiences in using first ARES and then ARES 2 in 4 courses.

Teaching of knowledge-intensive AI is particularly hard as the process
of how knowledge is acquired is difficult to grasp without practical experience. Acquiring
and using knowledge about actions, events, processes is especially difficult
because of the temporal nature of the subject matter. In this paper we report on a
tool called GIPO that has been used for teaching AI students the areas of knowledge
acquisition, knowledge engineering, automated planning and machine learning.
We give a short walkthrough of some of GIPO’s functions, indicating some
of the learning opportunities offered. We then compare GIPO with other interfaces
used in the computing curriculum. We argue that using a high level integrated tool
such as GIPO for supporting teaching and learning improves the students’ learning
experience, and helps integrate the theory and practice in a range of AI and related
subject areas.

We provide a summary of the lessons we have learned after teaching a graduate multiagent systems class over the last six years. The class has used various technologies such as RoboCup (along with our Biter and SoccerBeans tools), NetLogo, JADE, and FIPA-OS. We discuss their advantages and disadvantages. We also discuss our view of the future of multiagent systems. We notice the ongoing separation of software agent design from the theoretical underpinnings of multiagent theory and propose the development of a unifying notation for representing multiagent problems.