| Libellé du cours : | Neuromoprhic Technologies for Spiking Neural Networks |
|---|---|
| Département d'enseignement : | EEA / Electronique Electrotechnique Automatique |
| Responsable d'enseignement : | Monsieur PHILIPPE PERNOD |
| Langue d'enseignement : | |
| Ects potentiels : | 0 |
| Grille des résultats : | |
| Code et libellé (hp) : | MR_ETECH_S3_NTS - Neuromoprh Tech for Spiking N. |
Equipe pédagogique
Enseignants : Monsieur PHILIPPE PERNOD
Intervenants extérieurs (entreprise, recherche, enseignement secondaire) : divers enseignants vacataires
Résumé
The objective is to teach students Neuromorphic Technologies for Impulse Neural Networks (ANN), the keystone of "third generation artificial intelligence". Brief program: 1) Bioinspired information processing: - Nerve impulse in the living (characteristics of the brain, neurons, coding, biological membrane) 2) Artificial Neural Networks (ANN): brief history, architectures, software/hardware approaches, plasticity, supervised/unsupervised learning 3) Pulse Neuron Networks (PNN) - hardware implementations - for 3rd generation AI: (i) interest (response to the energy challenge, for which applications), description of the neuromorphic technologies (NT) used - all CMOS or integrating synapses from nanoelectronics - (ii) use of the PNNs in the development of AI 4) Coupling of NNS with bioinspired artificial sensors (retina, cochlea) 5) Bioinspired computing for hybrid biology / technology applications for information processing
Objectifs pédagogiques
Objectives (in terms of know-how): The objective is to teach students Neuromorphic Technologies for Impulse Neural Networks (SNN), the keystone of "third generation artificial intelligence". The student will understand the basic building blocks (neurons, synapses) required for the deployment of SNNs, acquire a culture related to neuromorphic technologies all CMOS or co-integrating synapses from nanoelectronics: organic or non-organic, magneto-electric. The coupling with bioinspired artificial sensors (retina, cochlea, ...) will be addressed. An opening to more exploratory directions aiming at reproducing the principles of information processing observed in biological systems using emerging technologies will also be proposed. This approach is particularly interested in reproducing intelligent sensor networks, exploiting the properties of complex systems at the nanoscale (i.e. reservoir computing) and exploring the coupling of electronics and biology for information processing. Acquired skills (direct/indirect): Consolidate the scientific culture of Master 2 students concerning neuromorphic technologies dedicated to 3rd generation AI
Objectifs de développement durable
Modalités de contrôle de connaissance
Contrôle Continu
Commentaires: -
Ressources en ligne
Simulation software for practical works
Pédagogie
Lectures & Tutorials : 16 Practical work: 12 Personnel work: 20
Séquencement / modalités d'apprentissage
| Nombre d'heures en CM (Cours Magistraux) : | 16 |
|---|---|
| Nombre d'heures en TD (Travaux Dirigés) : | 0 |
| Nombre d'heures en TP (Travaux Pratiques) : | 12 |
| Nombre d'heures en Séminaire : | 0 |
| Nombre d'heures en Demi-séminaire : | 0 |
| Nombre d'heures élèves en TEA (Travail En Autonomie) : | 0 |
| Nombre d'heures élèves en TNE (Travail Non Encadré) : | 0 |
| Nombre d'heures en CB (Contrôle Bloqué) : | 0 |
| Nombre d'heures élèves en PER (Travail PERsonnel) : | 0 |
| Nombre d'heures en Heures Projets : | 0 |
Pré-requis
Basics in Signal / Information processing
Nombre maximum d'inscrits
Remarques
This specific course is operated by University of Lille within the framework of the co-accreditation of the master between Centrale Lille and University of Lille.