Agents and Distributed Artificial Intelligence

Code:

EIC0033

Acronym:

AIAD

Keywords
Classification	Keyword
OFICIAL	Artificial Intelligence

Instance: 2014/2015 - 1S

Active?	Yes
Web Page:	http://paginas.fe.up.pt/~eol/AIAD/aiad1415.html
Responsible unit:	Department of Informatics Engineering
Course/CS Responsible:	Master in Informatics and Computing Engineering

Cycles of Study/Courses

Acronym	No. of Students	Study Plan	Curricular Years	Credits UCN	Credits ECTS	Contact hours	Total Time
MIEIC	126	Syllabus since 2009/2010	4	-	6	56	162

Teaching language

Suitable for English-speaking students

Objectives

We assume in this course a technological-oriented approach to Software Agents design and applications. Students are expected to acquire a technological perspective on the subject.

Agents Oriented Programming is introduced as a new metaphore for designing and implementing distributed computer systems. However, students will be able to deal with agents, as well as multi-agent systems, design through the support of formalization tools, including logics (intentional, BDI...). Through small projects, students will be able to illustrate agents and MAS concepts in their practical aspects and importance.

 

Learning outcomes and competences

At the end of the course, the student is expected to:

• Know the specificity of software agents

 

• Recognize and describe the classes of problems more appropriate to use Agents and Multi-Agent Systems

 

• Specify, by logical formalisms, the behavior of agents

 

• Explore tools for building Agents and Multi-Agent Systems

 

• Define and include intelligent decision strategies for Software Agents.

 

Working method

Presencial

Pre-requirements (prior knowledge) and co-requirements (common knowledge)

Knowledge of Artificial Intelligence techniques

Program

1. Distributed Artificial Imtelligence and Multi-Agent Systems

• Motivation and main Objectives

2. Agents

• Definitions, basic Architectures

• Knowledge Representation and Logic for Agents.

• Advanced Agents' Architectures. Subsumption and Reactive Agents; Mentalistic-like architectures and Deliberative Agents

• Learning Agents. Reinforcement Learning; Non-supervised Learning

3. Interaction in MAS

• Coordination and Cooperation. Strategies for Cooperation; Knowledge for Cooperation

• Supporting Communication. Agents communication languages: KQML and ACL; Ontologies: concepts, languages (XML, RDF),Tools; Platforms for agents communication: (JATLite), JADE, Brahms o Agents mobility (AGLETS)

4. Agents-Oriented Software Engineering. Improving GAIA methodology

5. Agents' Negotiation

• Contract Net and Market-based protocols

• Electronic Commerce .Open and Closed Auctions; MAS and Electronic Commerce; Learning strategies for trading

• Game Theory and Negotiation Domains. Concepts from Economics; Characterizing Negotiation Domains: TOD and WOD

• Negotiation techniques and Game Theory. Agents Joint Planning; Agreements, Coalitions and Utility measure

• Argumentation and Dialog Systems.

6. Emotion-like based Agent architectures.

7. MAS Application examples

• Modelo ARCHON

• Resources management application

• "Truth maintenance" Distributed System

• Electronic Institutions (ForEV)

• E-Brokering - BIAS

Mandatory literature

Eugénio Oliveira ; 'Cópias dos quadros tópicos das Aulas',
Michael Wooldridge; An introduction to multiagent systems. ISBN: 978-0-470-51946-2

Complementary Bibliography

Eds.M.Luck et al; Multi-Agent Systems and Applications, Springer, 2001
S. Russel and P. Norvig; 'Artificial Intelligence: A Modern Approach', Prentice Hall, 2003

Teaching methods and learning activities

Theoretical classes will be based on the oral presentation of the themes of the course, as well as interaction with students. Methods of implementation of applications will be taught (tools of specification and platforms for communication). Practical classes will be based on the supervision of students’ assignments. Reports are mandatory (in the middle and at the end of the semester). Project oriented learning.

Software

Plataforma de Sistemas Multi-Agente JADE
Ambiente de Simulação REPAST
Plataforma JADEX

keywords

Technological sciences > Engineering > Knowledge engineering

Evaluation Type

Distributed evaluation with final exam

Assessment Components

Designation	Weight (%)
Exame	50,00
Participação presencial	5,00
Trabalho escrito	25,00
Trabalho laboratorial	20,00
Total:	100,00

Amount of time allocated to each course unit

Designation	Time (hours)
Estudo autónomo	44,00
Frequência das aulas	56,00
Trabalho laboratorial	62,00
Total:	162,00

Eligibility for exams

Not exceed the absence limit allowed (25% of the Lab classes) and have a minimum of 35% of the possible total in the evaluation assignments (distributed classification - DC)

Calculation formula of final grade

FC = 0.5*DC + 0.5*EC

EC : Exam Classification (the use of pre-existing written material is allowed)

DC : Distributed Classification, includes:

- Intermediate Report plus demo: 15%

- Final Report:10%

- Project implementation and Demo: 20%

- Participation in class: 5%

To pass, the student must have a minimum of 35% in each of the two evaluation components, distributed and final exam.

Examinations or Special Assignments

One practical assignment and the respective report (weight=50%) and an exam (weight=50%).

To get approved, the student must have a grade equal or higher to 35% in each of the evaluation items.

Special assessment (TE, DA, ...)

The distributed evaluation is for all the students, regardless of their enrollment regime. Students enrolled under special arrangements without attending practical classes, must agree with teachers the work being undertaken and the dates of assessment (intermediate and final). These students are not subjected to the evaluation of "Participation in class"

Classification improvement

The classification improvement can be done by improving the exam AND/OR improving the project