Introduction to IntelligentAutonomous Systems
Keywords |
Classification |
Keyword |
OFICIAL |
Computer Science |
Instance: 2024/2025 - 1S
Cycles of Study/Courses
Acronym |
No. of Students |
Study Plan |
Curricular Years |
Credits UCN |
Credits ECTS |
Contact hours |
Total Time |
L:IACD |
56 |
study plan from 2021/22 |
3 |
- |
6 |
48 |
162 |
Teaching Staff - Responsibilities
Teaching language
Suitable for English-speaking students
Objectives
This course presents a global perspective of the techniques associated with intelligent and autonomous systems, exploring the modeling and simulation of complex systems and the development of intelligent agents and Multi-Agent Systems with the ability to adapt/learn to solve complex problems. The main objective is to specify and implement autonomous, complex, and adaptive intelligent systems. At the end of the course, students should be able to:
1. Understand basic concepts related to autonomous intelligent systems and be able to model and design complex intelligent and autonomous systems.
2. Understand and be capable of using concepts of intelligent multi-agent systems such as communication, interaction, coordination, negotiation, and cooperation.
3. Understand and be capable of using reinforcement learning, including state-of-the-art algorithms and deep reinforcement learning mechanisms.
Learning outcomes and competences
Ability to specify and implement autonomous, complex, and adaptive intelligent systems.
Working method
Presencial
Pre-requirements (prior knowledge) and co-requirements (common knowledge)
Knowledge of Python / Java programming and Artificial Intelligence.
Program
1. Intelligent Autonomous Systems: Intelligent Systems; Agents; Agent architectures.
2. Multi-Agent Systems. Agent-based simulation. Speech-act theory. Agent communication languages. Development platforms.
3. Multi-agent decision making. Game theory. Social choice theory. Mechanism design: protocols and auctions. Negotiation.
4. Learning in autonomous systems. Reinforcement Learning. Markov Decision Processes. Model-based and model-free methods. Monte Carlo. Temporal-difference learning: Q-learning, SARSA. Value function approximation. Policy gradient methods.
Mandatory literature
Michael Wooldridge;
An introduction to multiagent systems. ISBN: 0-471-49691-X
Richard S. Sutton, Andrew G. Barto; Reinforcement learning: an introduction (2nd edition), The MIT Press, 2018. ISBN: 978-0-262-19398-6 (http://incompleteideas.net/book/the-book-2nd.html)
Complementary Bibliography
Stuart Jonathan Russell;
Artificial intelligence. ISBN: 978-1-292-40113-3
Yoav Shoham, Kevin Leyton-Brown; Multiagent Systems – Algorithmic, Game-Theoretic, and Logical Foundations. ISBN: 978-0-521-89943-7 (http://www.masfoundations.org/)
Teaching methods and learning activities
Oral presentation of the themes of the course in theoretical classes, with interaction with students. Tool exploration and exercise solving in the practical classes. Practical classes will also be based on the supervision of assignments.
Software
SPADE
JADE
Evaluation Type
Distributed evaluation with final exam
Assessment Components
designation |
Weight (%) |
Exame |
50,00 |
Trabalho prático ou de projeto |
50,00 |
Total: |
100,00 |
Amount of time allocated to each course unit
designation |
Time (hours) |
Estudo autónomo |
50,00 |
Frequência das aulas |
48,00 |
Trabalho laboratorial |
64,00 |
Total: |
162,00 |
Eligibility for exams
Terms of frequency: Enrolled students must not exceed the allowed number of non-attendances to lab classes and achieve a minimum grade of 7.5 (out of 20) in each of the practical assignments.
Calculation formula of final grade
Final Grade = 0.35*TP1+0.15*TP2+0.5*E
TP1: practical assignment 1
TP2: practical assignment 2
E: exam
The minimum grade for the exam is 7.5 (out of 20).
Special assessment (TE, DA, ...)
Students who have been exempted from practical classes must agree with their teachers on how they will be accompanied in their practical work.
Evaluation at a special season corresponds to performing a practical assignment and the exam, each worth 50% in the final grade and keeping the minimum grades referred to in the normal assessment.
Classification improvement
The exam grade can be improved in the appeal period. The practical part cannot be improved.