Full course description
The course starts with an analysis of the question "Can machines think", and the preconceptions usually encountered in discussions about that idea.
Next the metaphor of an “intelligent agent” is introduced, that is, of an entity that pursues goals by perceiving and acting flexibly and autonomously in a possibly very complex environment.
The main part of the course explores the metaphor of an intelligent agent by introducing a number of state-of-the-art concepts, algorithms, and methods which enable computers (i.e., software and robots) to solve problems in a way which deserves to be called intelligent. Topics covered in this part are chosen from AI areas such as intelligent search and constraint satisfaction, architectures for intelligent agents, and coordination among intelligent agents.
The course as a whole conveys basic aspects and facets of engineering (analyzing and designing) AI systems. Covered topics are explored and applied in exercises and tasks (in-class and homework).
- To convey the ideas that have emerged over the past fifty years of Artificial Intelligence research, and about two millenia of related work, expressed in the study of so-called intelligent agents.
- To discuss the possibility of machines that think.
- To show how algorithms can be used (1) to understand human behavior in terms of underlying processes, and (2) to enable systems to think or act intelligently.
- Russell, S., & Norvig, P. (2009, Third Edition). Artificial Intelligence. A modern approach. Prentice-Hall.
- Wooldridge, M. (2009, Second Edition). An introduction to multiagent systems. John Wiley & Sons.