This course introduces students to the fundamental principles and practical techniques of machine learning with an overview of supervised, unsupervised, and generative learning paradigms. It begins by emphasizing hands-on experience by delving into Exploratory Data Analysis (EDA). It then develops and evaluates traditional supervised learning models for classification and regression. The course then covers the theoretical foundations of deep learning and the implementation/evaluation of Multilayer Perceptrons (MLPs) solutions for both classification and regression tasks. Learners will also explore clustering techniques, dimensionality reduction, and applications such as CNNs for Computer Vision (CV) and RNNs, LSTMs, and GRUs for Natural Language Processing (NLP). Students will engage with modern NLP tools including Hugging Face Transformers and explore pretrained models and annotation tools in CV. The course concludes with an introduction to Generative AI, including GANs, Diffusion Models, Attention Mechanisms, and Transformer architectures. All solutions are implemented in Python using industry-standard libraries such as Scikit-learn, TensorFlow, and Keras. Ethical and societal considerations—including fairness, bias, transparency, explainability, and privacy—are discussed to highlight the broader impact of machine learning technologies.

Temps présentiel : 30 heures

Charge de travail étudiant : 70 heures

Méthode(s) d'évaluation : Examen écrit, Travail personnel, Travaux pratiques