Código: | ERAGT052 | Sigla: | BAI |
Página Web: | http://arturmarques.com/edu/ia/ |
Área de Ensino: | Informática |
Sigla | Nº de Estudantes | Plano de Estudos | Ano Curricular | Créditos | Horas Contacto | Horas Totais |
---|---|---|---|---|---|---|
ERSGT | Curso Mobilidade Internacional Erasmus | 1º | 7,5 | 75 | 75 |
Teórico-Práticas: | 42,00 |
Docência - Horas Semanais
|
Docência - Responsabilidades
|
The approved student is expected to:
(o1) Understand the concept of "Artificial Intelligence" (AI)/"machine intelligence", its origins and currents, as well as being sensitive to the importance of responsible and ethical approaches;
(o2) Be aware of some of the well-established languages and tools for developing AI solutions;
(o3) Know how to apply the studied languages and tools, to the development of concrete solutions for specific problems, namely in the areas of search, knowledge and learning;
(o4) Have the ability to abstract available solutions and pre-trained models, but also be capable of understanding their results and underlying theory.
Introduction to Artificial Intelligence (AI)
- Concept
- Prominent areas
- History, milestones, case studies
- "Responsible" and "ethical" AI
Languages and Tools
- Selection
- Refresh/introduction to selected languages and tools
Search by intelligent agents
- Concepts and terminology: agent, state, initial state, actions, transition model, state space, objective(s), cost
- Programming of intelligent agent(s) for search
- Algorithms
Knowledge-based agents
- Concepts and terminology: logic, notation, operators, model, inference
- Programming with knowledge-based agents
Learning
- Concepts and terminology: classification, mappings, under/overfitting, supervised learning, transfer-learning, data pipeline, data augmentation
- Programming solutions for classifying inputs, using pre-trained models (transfer-learning); building models by different techniques, model assessment
The topics allow the student to walk a path that begins by discussing and understanding what "AI" is, the scope of the concept, and its multiple perspectives taken over the years, which have led to different areas of study. The computational power of the present and the transversal reach of "AI", demands a discussion on "responsible" and "ethical" AI (o1).
The practical development of solutions is done with languages and tools, which are to be selected and put into action (o2).
The selected languages and tools are applied to building software for specific search, knowledge and learning problems, gradually, as the related concepts are studied (o3).
This approach should translate in students capable of creating related solutions, including by reuse, with an understanding of the underlying fundamentals (o4).
Presentations and case studies.
Hands-on AI software development.
Assessment:
Projects proposed by the student and agreed with the teacher (T)
Assessment element under teacher¿s control (E)
Final grade = 0.4 * T + 0.6 * E
The presentations and the case studies introduce and build the concepts, helping in their understanding.
The hands-on development using the adopted solutions should translate to AI software creation skills.
Ertel, W. and N. T. Black (2017). Introduction to Artificial Intelligence, Springer.
Bird, A., et al. (2019). The Python Workshop: Learn to code in Python and kickstart your career in software development or data science Packt Publishing.
Ameisen, E. (2020). Building Machine Learning Powered Applications: Going from Idea to Product, O'Reilly Media.