-
Presentation
Presentation
This module is an introduction to the basic concepts and techniques of Artificial Intelligence, with three main focus areas. First, the formalisation of what a machine is, both in terms of the symbol manipulation of the Turing machine, and in the McCulloch and Pitts machines that work with interconnection patterns between nodes in neural networks. Second, the concept of rational agent in AI, that emerges the intersection with the Cognitive Sciences, and the various implementations of comprehensive structured search algorithms (informed and not informed). Still within this focus area, the concept of stochastic search and Constraint Satisfaction Problems (CSPs) are also introduced. Finally, in the third area of focus, students learn some notions and uses of some of the more advanced artificial intelligence algorithms that are used today.
-
Class from course
Class from course
-
Degree | Semesters | ECTS
Degree | Semesters | ECTS
Bachelor | Semestral | 5
-
Year | Nature | Language
Year | Nature | Language
3 | Mandatory | Português
-
Code
Code
ULHT12-2129
-
Prerequisites and corequisites
Prerequisites and corequisites
Not applicable
-
Professional Internship
Professional Internship
Não
-
Syllabus
Syllabus
Basic Concepts Definitions of AI Turing Machine McCulloch and Pitts Neural Networks How to Analyze Machines? State Transition Diagrams Search The Concept of a Search Agent in AI Spaces and Search Graphs Uninformed Search: British Museum, DFS, BFS Informed Search: Dijkstra and A* Adversarial Search Constraint Satisfaction Problems Basic Notions of natural language processing and recommendation systems The Future of AI Critical Analysis of Recent Articles in AI
-
Objectives
Objectives
The learning objectives of this course include (1) a deep understanding of the conceptual aspects that give rise to AI, namely the formalization of the concept of universal computing through symbol manipulation, and neural network-based computing; (2) the methods and representations used to study the functioning of any machine; (3) the design and implementation of rational agents and the concept of "information processing"; (4) classical algorithms of uninformed search: British Museum, DFS and BFS; (5) informed search: Dijkstra and A*; (6) the formalization and resolution of constraint satisfaction problems (CSP); (7) basic knowledge of advanced artificial intelligence techniques in the domains of machine learning and data science; and (8) knowledge about the uses of artificial intelligence in society including aspects related to ethics and the future of AI.
-
Teaching methodologies and assessment
Teaching methodologies and assessment
Innovative teaching methods are used to enhance learning. The flipped classroom approach is utilised, where students pre-learn the concepts, allowing class time to be dedicated to practical exercises, discussions, and projects. Interactive programming sessions via Jupyter Notebooks allow for experimentation with Python in a constant feedback environment. Project-based learning is essential, with students undertaking various projects that promote theoretical-practical skills and critical thinking. By incorporating these methods, the aim is to equip students with the technical skills and analytical mindset necessary to achieve a broad and current knowledge of the fundamentals and state of the art in artificial intelligence.
-
References
References
Russell, S., & Norvig, P. (2020). Artificial intelligence: a modern approach. 4th edition. http://aima.cs.berkeley.edu
-
Office Hours
Office Hours
-
Mobility
Mobility
No
