BME VIK - Kooperáció és intelligencia

A tantárgy neve magyarul / Name of the subject in Hungarian: Kooperáció és intelligencia

3. Course coordinator and department Dr. Dobrowiecki Tadeusz Pawel,

4. Instructors Dr. Tadeusz Dobrowiecki, associate professor Department of Measurement and Information Systems

5. Required knowledge Artificial intelligence, Programming techniques (Java), Cooperative and learning systems

6. Pre-requisites

Ajánlott:
Artificial intelligence (BSc), Cooperative and learning systems (BSc).

7. Objectives, learning outcomes and obtained knowledge The aim of the subjects is multiple level analysis of cooperation, from basic informational structures to the solutions used in intelligent systems. During the semester, starting from low-level infrastructural layers, then progressing toward higher-level layers calling for intelligent behavior, we deal with the following topics: distributed machine learning, game theoretical problems, voting systems, ontologies, languages, communication and cooperation, distributed problem solving via cooperative communication. We expect that the students successfully fulfilling the requirements of the subject will have a clear view of the potential of the cooperative system solutions and of the spectrum of the cooperative technologies, will be able to design and analyze cooperative system models in practical applications, will gain working knowledge of the cooperative game theoretical solutions and high-level AI methods used in cooperative intelligent systems.

8. Synopsis I. Agent basics: Basic topics in communication and cooperation. Agents. Intelligent agents. Multi-agent systems.

II. Logical basis of agent communication and beyond: Cooperation in agent systems based on modal logical models (Belief-Desire-Intention). BDI model based agent modeling and communication languages - AgentSpeak(L) and Jason platform.

III. Learning in multi-agent systems: Special problems and opportunities of learning in multi-agent systems. Learning in cooperative organizations. Learning in competitive organizations. Learning in hierarchical organizations.

IV. Conceptual systems and cooperation: Basics of ontological systems. The role of ontologies. Ontologies in agent communication, in open systems. Ontologies in agent-human interactions.

V. Game theoretical models: Cooperation and Game Theory. Utility Theory: preference, utility, transitivity. Non-cooperative games. Rationality, common knowledge, perfectness, completeness, players, pure and mixed strategies, Nash-equilibrium, types, incomplete information (Bayesian) games, cooperative games.

VI. Competition in open systems: Cooperative conflict resolution. Auction- and Voting Theory. Single/multi-item, first/second-price, and sealed-bid auctions. Mechanism design, Social choice functions, dominant- and Nash-implementation, Vickrey-Clarke-Groves mechanisms, Revelation principle.

VII. Planned activities: Multi-agent planning. Basics. Planning in open systems with incomplete information. Embedding plans into agent modeling languages. Planning and communication planning.

Practical knowledge related to the subject is presented within the Cooperation and machine learning Lab.

9. Method of instruction
Theoretical part of the curriculum is taught during the lectures. Practical experimentation is supported by the home work, software demonstrations during the lectures, and by the related laboratory practice.

10. Assessment a. During the semester:

· 8 small home assignments to be handed out every week and a large home assignement. Reports with the solutions are due on the 13^th week of the semester. The joint presentation of the solutions and the qualification (assuming the reports are ready) is on the 14^th week of the semester. The assignment brings max. 40 points, the required minimum is 40%. Small assignments bring 0 … 5 points (bad, good, very good). The required minimum is 5 points.

b. During the examination period: oral exam. Students qualify for the final exam with minimal level achievements in home assignments (40 %).

c. Qualification: The final mark is based on the number of points collected from the home work, and the final exam.

11. Recaps Failed home assignments can be handed in until the end of the supplementary week.

12. Consultations As required, upon agreement.

13. References, textbooks and resources Lecture notes made available at the home page of the subject, suggested electronic literature and additional information, and a web link collection. Wooldridge, M., An Introduction to Multi-agent Systems, J. Wiley, 2002

Rafael H. Bordini, Jomi Fred Hübner, Michael Wooldridge, Programming Multi-Agent Systems in AgentSpeak using Jason, J. Wiley, 2007

Stuart Russell and Peter Norvig: Artificial intelligence. The modern approach, 2^nd edition, Prentice Hall, 2001 T. Mitchell: Machine Learning, McGraw-Hill, 1997. F. L. Bellifemine, G. Caire, D. Greenwood: Developing Multi Agent Systems with JADE, Wiley, 2007

14. Required learning hours and assignment

Kontakt óra	42
Félévközi készülés órákra	7
Felkészülés zárthelyire	0
Házi feladat elkészítése	15
Kijelölt írásos tananyag elsajátítása	8
Vizsgafelkészülés	48
Összesen	120

15. Syllabus prepared by

Dr. Tadeusz Dobrowiecki

Tamás Mészáros

Dániel László Kovács

associate professor

senior lecturer

assistant lecturer

DMIT