- L'Université de Lyon
- Formation
-
Recherche
-
Innovation et entrepreneuriat
- Actualités et présentation
- Entreprendre pendant et après ses études : le Centre d'Entrepreneuriat
- Innover autrement : Fabrique de l’Innovation
- Accélérer les innovations : PULSALYS
- Coordonner l'innovation à potentiel scientifique : PUI IMPULSE
- Entreprendre et innover avec un doctorat
- Découvrir l’industrie : INDULO, un dispositif de médiation industrielle
- International
-
Vie des campus
- Actualités et présentation
-
- Carte des campus
- Pilotage immobilier
-
Responsabilité sociétale et environnementale
- Climat comment agir ?
- La stratégie de site
- Stratégies et études prospectives
-
Culture, sciences et société
- Actualités et présentation
-
- Le portail Pop'Sciences
- La programmation toute l'année
- Le Pop'Sciences Mag
- Festival Pop'Sciences
-
Pop'Sciences Jeunes
- Pop'Sciences Jeunes reportages 2023/2024
- Stage 3e
- Pop'Sciences et le projet LYSiERES²
- Fête de la science
-
Partager cette page
Emploi
PhD position: "When computational physics meets observations: using machine learning to bridge the gap" - LGL-TPE (UMR5276) et CRAL(UMR5574)
CDD - 3 ansVilleurbanne, France
Date limite de réponse : 1 mai 2021
Missions :
Short description
The PhD position is proposed for a 3-year period (36 months). The legal gross salary is €1768 per month (plus social benefits). An annual €2 000 package for travels and equipment will be allotted. The candidate is expected to submit a thesis manuscript to the university of Lyon for a formal presentation in front of a jury before the end of the 3-yr period.
Starting date of the contract: October the 1st, 2021Research project
ContextNatural sciences such as astrophysics, geophysics and nuclear physics often use numerical simulations to model highly complex physical systems. These simulations are now more and more accurate thanks to the computational power available. For example, 3D convection models can simulate the thermochemical evolution and structure of stars and planets.
However, to disentangle different models and to estimate physical parameters (e.g., initial conditions), the outputs of these simulations have to be compared to observations. This confrontation of simulations to observations is a major challenge in natural sciences. Indeed, numerical simulations are now able to model quite accurately objects that are impossible to observe directly (e.g., interior of stars and planets, stars and black holes environment …). As for the observations, although their quality and quantity are rapidly increasing, they are often only indirectly related to the actual parameters of interest (e.g., seismic waves observations are used to construct images of the earth mantle, measured interferometric visibilities are used to characterize planet forming disk …).
To infer simulation parameters from observations is very challenging. When a single simulation is computationally intensive, it is impossible to use either stochastic or continuous optimization methods to infer parameters. In most cases, one can only rely on finding the best fits on a low dimensional pre-computed grid of model parameters.
Objectives
The ultimate goal of the proposed thesis is to build a fast interpolation method on a grid of computational physics simulated images (in a broad sense as it can also be 3D volumes or spectra). With such a method, we will quickly have an approximation of a simulated image from any possible set of parameters, without having to run the expensive simulation. It then will be possible to use any method (optimization, Bayesian inference) to derive the so sought-after distribution of parameters.
The main idea is to use a deep learning framework to build the interpolator. Indeed, all possible modeled images are concentrated on a lower-dimensional subspace or manifold. Deep neural networks such as Generative Adversarial Networks (GAN) [1] appear to be very efficient to model manifolds and could be much more efficient interpolators than classical polynomial interpolators. Trained on a grid on simulated images, these deep neural networks will produce continuous approximations of the images. As a toy example, in a properly defined manifold, the images of a single circle vary continuously with the circle radius. Interpolation between two images of circles with different radius must follow this manifold whereas any polynomial interpolation will produce an image with two circles rather than an image of a single circle with intermediate radius.
Grids of models are quite ubiquitous in physics, and hence such a project can have important impact. To ensure that it will be both robust and useful in practice, the deep learning based interpolator will be developed for two different applications: (i) planet forming disk characterization using VLTI in collaboration with J. Kluska (KU Leuven) and (ii) reconstruction of mantle structure based on geophysical surface observations.
Starting date of the contract: October the 1st, 2021
PROFIL RECHERCHÉ
Formations requises :
The candidates must hold a national master degree or equivalent.
Compétences requises :
A background in maths, physics, and programing is expected. We look for candidates with strong comunications skills, as the PhD will be carried out between different labs and different fields.
Informations complémentaires
SELECTION PROCESS
The successful candidate will be selected in partnership with the Doctoral School « Physics and Astrophysics » of the University of Lyon.Application deadline
May the 1st, 2021
Candidates on the short list will be informed by the end of May. They will be interviewed in June.
Renseignements pratiques
Informations sur l'organisme
LGL-TPE (UMR 5276) et CRAL (UMR 5574)
Adresse : LGL-TPE (UMR 5276)
69622 - Villeurbanne, France
69622 - Villeurbanne, France
Job location and description
Being co-supervised, the thesis student will be hosted either in the LGL Geosciences laboratory located on the La Doua Campus of Lyon 1 University (Villeurbanne) or in CRAL Astrophysic laboratory in Lyon Observatory (St Genis Laval) depending on the practical direction followed by the work.
Team
CRAL: Team HARISSA (High Angular Resolution, Imaging science, and Stellar Surroundings Astrophysics). Ferréol Soulez already co-supervised a student at laboratoire Hubert Curien in St Etienne.
Being co-supervised, the thesis student will be hosted either in the LGL Geosciences laboratory located on the La Doua Campus of Lyon 1 University (Villeurbanne) or in CRAL Astrophysic laboratory in Lyon Observatory (St Genis Laval) depending on the practical direction followed by the work.
Team
CRAL: Team HARISSA (High Angular Resolution, Imaging science, and Stellar Surroundings Astrophysics). Ferréol Soulez already co-supervised a student at laboratoire Hubert Curien in St Etienne.