Positionnement de l’UE : • Place de l’UE dans le cursus : Master • UE obligatoire/optionnelle: Lien avec le référentiel de compétences / les résultats d’apprentissage programme: The goal of this course is to give learners basic understanding of modern neural networks and their main applications in computer vision (image recognition) and natural language understanding (NLP). Students will study all popular building blocks of neural networks including fully connected layers, convolutional and recurrent layers. The course will prepare students for research and / or industrial application of deep learning techniques. Objectif général ou finalité : Data Science 1 course was based on building traditional machine learning models. These models rely on input features that may come from a domain expert. Data Science 2 course focuses on Deep Learning models and their implementation using different Neural Networks architectures (like DNN, CNN, RNN). Neural networks are able to automatically learn the features from data that are most useful for a particular task, like extracting automatically features from photos to classify them. The main goal of this course is learn the different architecture of NN where each one is useful for a type of problems. More specifically students will learn (1) Deep Neural Networks to build models that can predict hand-written digits; (2) Convolutional Neural Networks to recognize objects from photos; (3) Recurrent Neural Networks to apply it on sequential data, like predict the class (sentiment) of a text (since a text is sequential data).

Temps présentiel : 17.5 heures

Charge de travail étudiant : 82.5 heures

Méthode(s) d'évaluation : Examen final, Participation et assiduité, Projets

Référence :
Raschka, Sebastian, and Vahid Mirjalili. "Python Machine Learning: Machine Learning and Deep Learning with Python." scikit-learn, and TensorFlow. Packt Publishing, 2017. Heaton, Jeff. "Ian goodfellow, yoshua bengio, and aaron courville: Deep learning." (2018): 305-307.