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

TITLE OF THE POSITION : (M/F) Random Graphs in Machine Learning

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

Date Limite Candidature : mercredi 13 juillet 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 : UMR5216-VIRFAU-028
Date of publication : Wednesday, June 22, 2022
Scientific Responsible name : KERIVEN Nicolas
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 October 2022
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly

Description of the thesis topic

In the last decade, machine learning (ML) on graph data has known a rapid growth, with the advent of kernel methods on graphs, Graph Neural Networks (GNNs), benchmark datasets, and numerous applications ranging from the analysis of social networks to molecular classification or protein interface prediction.

Despite this, the study of traditional ML guarantees such as generalization bounds or sample complexities, which characterize the amount of training data needed to guarantee a low prediction error on new test data, has remained limited in the literature. This is often due to the lack statistical modelling of the graph structure: indeed, when no assumption is made on the graph-generating process to characterize unseen test data, the very notion of generalization becomes ill-defined.

On the other hand, Random Graph (RG) models represent a vast field of study in statistics and graph theory, but have hardly been explored in ML, despite immediate connections. In particular, Latent Position Models (LPM: each nodes is associated to an unknown latent variable, and edges are randomly generated according to these variables) such as Stochastic Block Models, graphons or epsilon-graphs, offer striking similarities with classical ML settings, with the significant difference that latent variables are unobserved and must be indirectly deduced from the graph structure.

The goal of this thesis is to explore the use of RG models in ML, and to study quantities such as generalization bounds, sample complexities, and so on, in this context. We will compare several ML models including GNNs, and, depending on the progress of the candidate, several RG models, such as LPMs and preferential attachment models. The internship will balance between theoretical studies and validation on real data, depending on the candidate.

Work Context

Gipsa-lab is a CNRS research unit joint, Grenoble-INP (Grenoble Institute of Technology), University of Grenoble under agreement with Inria, Observatory of Sciences of the Universe of Grenoble.
With 350 people including about 130 doctoral students, Gipsa-lab is a multidisciplinary research unit developing both basic and applied researches on complex signals and systems.
Gipsa-lab develops projects in the strategic areas of energy, environment, communication, intelligent systems, life and health and linguistic engineering.
Thanks to the research activities, Gipsa-lab maintains a constant link with the economic environment through a strong partnership with companies.
Gipsa-lab staff is involved in teaching and training in the various universities and engineering schools of the Grenoble academic area (Université Grenoble Alpes).
Gipsa-lab is internationally recognized for the research achieved in Automatic & Diagnostics, Signal Image Information Data Sciences, Speech and Cognition. The research unit develops projects in 16 teams organized in 4 Reseach centers
.Automatic & Diagnostic
.Data Science
.Geometry, Learning, Information and Algorithms
.Speech -cognition

Gipsa-lab regroups 148 permanent staff and around 260 no-permanent staff (Phd, post-dotoral students, visiting scholars, trainees in master…)

Constraints and risks


Additional Information


We talk about it on Twitter!