en

PhD position When computational physics meets observations: using machine learning to bridge the gap

PhD position When computational physics meets observations: using machine learning to bridge the gap

France 01 May 2021
LabEx LIO

LabEx LIO

State University (France), Browse similar opportunities

OPPORTUNITY DETAILS

Total reward
0 $
State University
Area
Host Country
Deadline
01 May 2021
Study level
Opportunity type
PhD
Specialities
Opportunity funding
Not funding
Eligible Countries
This opportunity is destined for all countries
Eligible Region
All Regions

Short description

The PhD position is proposed for a 3-year period (36 months). The legal net 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, 2021

Research project

Context

Natural 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.

Research field(s)

Machine learning, inverse problems and signal processing

Thesis supervisor and contact

Name: T. Bodin

Laboratory: LGL-TPE

Phone number: +33 (0)4 72 44 79 91

Email: thomas.bodin@ens-lyon.fr

The thesis will be co-supervised by

Name: F. Soulez

Laboratory: CRAL

Phone number: +33 (0)4 78 86 85 46

Email: ferreol.soulez@univ-lyon1.fr

Working environment

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

Allocated resources

The student will have access to the super computer of the seismology team at LGLTPE.

Recent publications of the team

A. Berdeu, F. Soulez, L. Denis, M. Langlois, É. Thiébaut,

PIC: a data reduction algorithm for integral field spectrographs

Astronomy and Astrophysics - A&A, EDP Sciences, 2020, 635, pp.A90.

J. Kluska, J.-P. Berger, Fabien Malbet, B. Lazareff, M. Benisty, J.-B. Le Bouquin, O. Absil, F. Baron, A. Delboulbé, G. Duvert, A. Isella, L. Jocou, A. Juhasz, S. Kraus, R. Lachaume, F. Ménard, R. Millan-Gabet, J. D. Monnier, T. Moulin, K. Perraut, S. Rochat, C. Pinte, F. Soulez, M. Tallon, W.-F. Thi, E. Thiébaut, W. Traub, G. Zins

A family portrait of disk inner rims around Herbig Ae/Be stars. Hunting for warps, rings, self shadowing, and misalignments in the inner astronomical units

Astronomy and Astrophysics - A&A, EDP Sciences, 2020,

E. Soubies, F. Soulez, M. T. Mccann, T-A. Pham, L. Donati, T. Debarre, D. Sage, M.Unser

Pocket Guide to Solve Inverse Problems with GlobalBioIm

Inverse Problems, IOP Publishing, 2019, 35 (10), pp.104006.

Seismic evidence for partial melt below tectonic plates

E Debayle, T Bodin, S Durand, Y Ricard

Nature 586 (7830), 555-559

Quantifying seismic anisotropy induced by small-scale chemical heterogeneities

C Alder, T Bodin, Y Ricard, Y Capdeville, E Debayle, JP Montagner

Geophysical Journal International 211 (3), 1585-1600

Description of LabEx LIO

In 2011, The Lyon Institute of Origins LabEx was selected following the first “Laboratory of Excellence” call for projects, part of the “Investissement d’Avenir” program for forward-looking research. It is one of 12 LabExes supported by the University of Lyon community of universities and establishments (COMUE). LIO brings together more than 200 elite researchers recruited throughout the word and forming 18 research teams from four laboratories in the Rhône-Alps region, all leaders in their fields, under the auspices of the University Claude Bernard Lyon 1 (UCBL), the Ecole Normale Supérieure de Lyon, and the CNRS. LIO’s goal is to explore questions about our origins, operating in a broad field of study that ranges from particle physics to geophysics, and includes cosmology, astrophysics, planetology and life.

Selection process


The successful candidate will be selected in partnership with the Doctoral School « Physics and Astrophysics » of the University of Lyon.

Condition for admission to doctoral studies

The candidates must hold a national master degree or equivalent.

Application deadline

May the 1st, 2021

Requested documents for application

The candidates must submit their application with (i) their academic curriculum of the last three years, (ii) a letter of motivation, (iii) a CV and (iv) a letter of recommendation, to labex.lio@universite-lyon.fr before May the 1st, 2021.

Candidates on the short list will be informed by the end of May. They will be interviewed in June.

Other organizations


Choose your study destination


Choose the country you wish to travel to study for free, work or volunteer

Please find also