UE Artificial intelligence : Introduction to deep learning

Code de l'ECUE : KMUIAIUI

Ce cours est proposé dans 8 UE

Structure

EUR DS4H

Domaine disciplinaire

Informatique

Lieu d'enseignement

Campus SophiaTech Les Templiers

Niveau du cours

Master 1 , Master 2 , Doctorat

Semestre

Semestre pair

Langue

Anglais

PRESENTATION

This course introduces the key concepts, techniques, and real-world applications of deep learning. Designed for learners with prior knowledge of traditional machine learning, it provides the theoretical background and hands-on experience needed to build and optimize neural networks for complex AI tasks.

You will learn about the evolution of deep learning, core architectures such as CNNs and RNNs, optimization techniques (Dropout, Batch Normalization), and practical implementation using Python, TensorFlow, Keras, and Scikit-Learn. Applications include image recognition, NLP, and other AI-driven solutions.

Prerequisites: Basic experience in Python and familiarity with NumPy, Pandas, and Matplotlib.

By the end of the course, you will be able to design and implement neural networks for a variety of AI applications.

Placement Test (Recommended)

To help you start with confidence, we recommend completing a short placement test before enrolling. The test takes about 10 minutes, is not an exam, includes no deep learning questions, and has no impact on enrollment or certification.

Link to the placement test

Responsable(s) du cours

Federica Granese , Anais Ollagnier

Présentiel

12.5h de cours magistral
16h de travaux dirigés

PREREQUIS

Avant le début du cours, je dois ...

Prior knowledge of basic Python programming
Familiarity with machine learning concepts (e.g., supervised learning, classification, regression). Beginners may consult Hal Daumé’s A Course in Machine Learning (Chapters 1–5 recommended), The EFELIA Côte d'azur Handbook on AI.
An introductory understanding of probability, statistics, and linear algebra is helpful but not mandatory.
The main scientific libraries : NumPy, Pandas, and Matplotlib

OBJECTIFS

A la fin de ce cours, je devrais être capable de...

Acquire the ability to design and implement deep learning models for a range of real-world applications
Understand the fundamentals of deep learning.
Develop a strong understanding of the theoretical foundations and practical challenges of modern neural networks, including their potential and limitations.
Know how to build neural network-based models to process structured data, images, or text.
Gain proficiency in PyTorch, Keras 3, and JupyterLab.

CONTENU

Introduction to Deep Learning

Aucune description
Neuronal Networks and Multilayer Perceptrons

Aucune description
Convolutional Neuronal Networks (CNNs)

Aucune description
Encoder/Decoder Networks

Aucune description
Data, Embedding, and Latent Spaces

Aucune description
Recurrent Neural Networks (RNNs) and Transformers

Aucune description
Large Language Models (LLMs)

Aucune description
Learning Optimization (params & metrics)

Aucune description

Accéder au Syllabus complet (Authentification requise)

Important

Ce syllabus n’a aucune valeur contractuelle. Son contenu est susceptible d’évoluer en cours d’année : soyez attentifs aux dernières modifications.