En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)

Large-scale spiking neural network simulations and their applications (M/F)

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : mercredi 12 octobre 2022

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

General information

Reference : UMR5549-SIMTHO2-002
Workplace : TOULOUSE
Date of publication : Wednesday, September 21, 2022
Scientific Responsible name : Simon THORPE DRCE2 CNRS CerCo et Emiliano LORINI CR CNRS IRIT
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 31 October 2022
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly

Description of the thesis topic

The main objective of the thesis will be the development of a simulator capable of simulating billions of spiking neurons using very sparse, binary connections. We will implement learning algorithms that also use binary connections, allowing neurons to learn to respond selectively to repeating patterns of activity. The initial implementations will use purely feedforward processing architectures, but later versions will allow fully recurrent networks to be implemented. We will compare performance on a range of currently available hardware platforms, including NVIDIA GPUs and Apple M1 silicon. Later in the project, we will explore the network's performance in various application areas, including text processing, knowledge representation and causal modelling. Finally, the thesis will also contribute to WP3 of the ANR project, namely, the development of a hybrid architecture in which the Spiking Neural Network is used to implement planning decisions.

Work Context

The thesis will be carried out in the context of an ANR-funded project, ALoRS "Action, Logical Reasoning and Spiking networks" and involves a collaboration between the Brain & Cognition Research Center (CerCo, UMR-5549) and the Institut de Recherche en Informatique de Toulouse (IRIT, UMR-5505) that started on the 1st of April 2022. The current project overlaps strongly with WP1 of the ANR project, designed to work on the "Study and Implementation of a Spiking Neural Network" and involves several subtasks.

Constraints and risks

Most of the work will involve programming computers and has no particular risks associated.

We talk about it on Twitter!