Fondations de l'intelligence artificielle

This course introduces students to the foundations of Artificial Intelligence. Seven main themes are covered:

• Introduction to AI: where we stand today, the difference between symbolic and connectionism AI, what truly distinguishes AI, ML, and DL, with concrete examples drawn from your own fields (physics, chemistry, biology...), and much more.
• Data and preprocessing: because no model is ever better than the data it's given: how to explore, clean, transform, normalize data, and select the right variables.
• The major families of learning: supervised, unsupervised, reinforcement learning: understanding what sets them apart and when to use each.
• Classical algorithms: regression, classification, decision trees, SVM, clustering: the basic toolbox of machine learning.
• Evaluating and validating a model: overfitting, cross-validation, metrics: how to know if a model is genuinely good, or just good on paper.
• Deep Learning and computer vision: neural networks, convolutional and recurrent architectures, transfer learning, and their applications.
• Natural language processing: from language models to Transformers, up to large language models (LLMs) and the art of prompting.

By the end of this course, students will have the knowledge and technical skills needed to work on AI-related projects and to implement its methods and tools within their respective fields. They will also be able to critically analyze the strengths and limitations of the models they develop and assess their relevance within their application context.

Required and Corequired knowledge and skills

Basic programming, elementary algorithmics, basic mathematics (linear algebra and probability).

Required and corequired courses

• Lectures: theoretical presentations illustrated with concrete examples and demonstrations.
• Guided practical exercises: applying concepts to simple datasets, and implementing models using standard libraries (Python).
• Applied project: individual or group work on a dataset chosen by the student.
• Guided self-study: readings and supplementary resources between sessions.

References, bibliography, and recommended reading

The books listed below serve as supplementary references for students who wish to deepen their knowledge:

• Hands-on Machine Learning with Scikit-Learn, Keras and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems. (2019). 2nd ed. CA 95472: O’Reilly.
• Russell, S., & Norvig, P. (2021). Artificial Intelligence: A Modern Approach (4th ed.). Pearson.
• Zheng, A., & Casari, A. (2018). Feature engineering for machine learning: Principles and techniques for data scientists. O'Reilly Media
• Reis, J., & Housley, M. (2022). Fundamentals of data engineering: Plan and build robust data systems. O'Reilly Media. 

Other resources, including links to courses, tutorials, and supplementary teaching materials, will also be shared directly in the slides, allowing students to explore certain topics in greater depth and further their learning.

Course notes

• Université virtuelle

Additional information

I will ensure that course slides are as complete as possible, including, where relevant, links to interesting references as well as supplementary slides to review and study at home. Materials will be made available on UV-ULB after each session.

Contacts

ihsane.gryech@ulb.be

Campus

Plaine

Method(s) of evaluation

• written examination
• Project
• Continuous assessment

written examination

Project

Continuous assessment

Evaluation consists of three parts:

1. Written exam: assesses theoretical knowledge and understanding of the algorithms covered.
2. Project: assesses the ability to apply course concepts to real data - implementation, testing, and analysis of a complete AI pipeline.
3. In-class quizzes and homework: assess students' progressive understanding through regular exercises throughout the semester.

Mark calculation method (including weighting of intermediary marks)

• 50% written exam
• 30% project
• 20% in-class quizzes and homework

Language(s) of evaluation

• french
• (if applicable english )