Distributed and Federated Learning
Code UE : USEES3
- Cours
- 5 crédits
Responsable(s)
Stefano SECCI
Public, conditions d’accès et prérequis
- Students are required to have taken an introductory machine-learning course
- Good knowledge on supervised learning.
- Some knowledge on Gradient descent
- Bases on unsupervised learning is recommended, but this is not a prerequisite.
Objectifs pédagogiques
This course provides an overview of federated and distributed learning in terms of performance and sécurité. Both theoretical and practical aspects will be extensively explored in this course in order to acquire solid expertise on both aspects. By the end of the course, students should
- Understand the major difference between centralized and decentralized learning
- Understand the methods most commonly used in federated learning
- Understand the performance of federated learning when data is heterogeneous (Non-IID data).
- Be able to build and scale a simple federated system
- Have acquired competence in implementing federated learning in the fields of networks, security, health or others.
Contenu
This course is designed to extend students' knowledge of learning in a decentralized setting. Decentralized learning techniques, such as federated learning, are set to deliver a new generation of machine learning applications by enabling efficient and reliable learning between multiple parties and from diverse data sources. This course will cover different aspects of federated learning, focusing on recent research developments and exploring important applications in different fields such as security, networks and healthcare.
Lectures
Lectures
- Course Overview. Introduction to machine learning and Federated Learning.
- Decentralized Optimization and Gradient descent
- Federated learning: FedSGD and FedAvg
- Variations of Federated Aggregation.
- Federated Averaging with Heterogeneous Data
- Communication-Efficient Learning of deep networks in Federated Learning
- Federated Multi-Task learning
- Build and scale a simple federated learning with MNIST, Cifar-10, Fashion-MNIST, MedMNIST, Shakespeare, and BCN Open Data. Open-source Federated Learning tools (Pytorch, Flower, etc.).
- Federated learning with Non-IID data.
- Threats, attacks, and defenses to federated learning
- Designing an attack and setting up a defense for federated learning.
- Applications to Images, Networks, health, and vehicle-to-vehicle communications
- Applications of federated learning to network anomaly detection: use of 5G and LoRaWAN testbeds and datasets, with lab [by CNAM].
- Applications of federation learning to medical equipment: use of aggregated and anonymized field data [by NTUU].
- Applications of federation learning to vehicle-to-vehicle communications: routing and content offloading [by UPC].
Modalité d'évaluation
- The evaluation will be based on a final exam, lab reports and/or project activity.
- For the project, students may also conduct research in the field of federated learning and write a short paper.
Bibliographie
- Kiyoshi Nakayama, George Jeno : Federated Learning with Python, O’Reilly, October 2022
- Lam M. Nguyen, Trong Nghia Hoang, Pin-Yu Chen : Federated Learning.Theory and Practice. Elsevier 2024. ISBN: 9780443190384
Cette UE apparaît dans les diplômes et certificats suivants
Rechercher une formation
RECHERCHE MULTI-CRITERES
Plus de critères de recherche sont proposés:
-
Vous pouvez sélectionner des formations grâce à un mot ou à une expression (chaîne de caractères) présent dans l’intitulé de la formation, sa description ou ses index (discipline ou métier).
Des mots-clés sont suggérés à partir du 3e caractère saisi, mais vous pouvez aussi rechercher librement. - Les différents items sélectionnés sont croisés.
ex: "Comptabilité" et "Diplôme" - Les résultats comprennent des formations du Cnam Liban (UE, diplômes, certificats, stages) et des formations proposées à distance par d'autres centres du Cnam.
- Les codes des formations du Liban se terminent par le suffixe LIB.
- Dans tous les cas, veillez à ne pas insérer d'espace ni de ponctuation supplémentaire.
Plus de critères de recherche sont proposés:
- Type de diplôme
- Niveau d'entrée
- Modalité de l'enseignement
- Programmation semestrielle
Chargement du résultat...
Intitulé de la formation |
Type |
Modalité(s) |
Lieu(x) |
|
---|---|---|---|---|
Intitulé de la formation
Artificial Intelligence for Connected Industries
|
Lieu(x)
Package
|
Lieu(x)
Paris
|
||
Intitulé de la formation
Master ROC en alternance - Mulhouse
|
||||
Intitulé de la formation | Type | Modalité(s) | Lieu(x) |
Contact
Voir le calendrier, le tarif, les conditions d'accessibilité et les modalités d'inscription dans le(s) centre(s) d'enseignement qui propose(nt) cette formation.
Enseignement non encore programmé
Code UE : USEES3
- Cours
- 5 crédits
Responsable(s)
Stefano SECCI