Feedback

Faculté des Sciences appliquées
Faculté des Sciences appliquées
MASTER THESIS
VIEW 172 | DOWNLOAD 573

Master's Thesis : Lightning Gravitational Wave Parameter Inference through Neural Amortization

Download
Delaunoy, Arnaud ULiège
Promotor(s) : Louppe, Gilles ULiège
Date of defense : 7-Sep-2020/9-Sep-2020 • Permalink : http://hdl.handle.net/2268.2/10554
Details
Title : Master's Thesis : Lightning Gravitational Wave Parameter Inference through Neural Amortization
Author : Delaunoy, Arnaud ULiège
Date of defense  : 7-Sep-2020/9-Sep-2020
Advisor(s) : Louppe, Gilles ULiège
Committee's member(s) : Geurts, Pierre ULiège
Wehenkel, Louis ULiège
Hermans, Joeri ULiège
Language : English
Number of pages : 63
Keywords : [en] Gravitational wave
[en] Deep learning
[en] Machine learning
[en] Simulation-based inference
[en] Fast inference
Discipline(s) : Engineering, computing & technology > Computer science
Physical, chemical, mathematical & earth Sciences > Physics
Research unit : Montefiore Institute, University of Liège
GRAPPA Institute, University of Amsterdam
Name of the research project : Lightning Gravitational Wave Parameter Inference through Neural Amortization
Target public : Researchers
Professionals of domain
Institution(s) : Université de Liège, Liège, Belgique
Degree: Master : ingénieur civil en science des données, à finalité spécialisée
Faculty: Master thesis of the Faculté des Sciences appliquées

Abstract

[en] Gravitational waves analysis relies on a simulator governed by the nonlinear field equations of general relativity for binary systems. Such analysis is computationally very expensive and necessitates a large-scale exploration of the likelihood surface over the full parameter space.
Neural networks have been gaining popularity as tools for gravitational waves analysis for the last few years. They lead to fast gravitational wave detection and parameter inference and hence complement classical slower techniques. This work explores simulation based inference which relies on likelihood-to-evidence ratio estimation for the parameters of binary black holes mergers. We build a neural network modeling this ratio and use it in place of the simulator allowing to perform parameter inference in few minutes. The performances are assessed on both gravitational wave generated by the simulator and emitted by real black holes.


File(s)

Document(s)

File
Access thesis.pdf
Description:
Size: 3.85 MB
Format: Adobe PDF

Annexe(s)

File
Access summary.pdf
Description:
Size: 134.71 kB
Format: Adobe PDF
File
Access mass_mocks.pdf
Description: Examples of inference of the masses performed on simulated gravitational waves. The 50% and 90% credible intervals are derived. The blue dot represents the maximum a posteriori estimator and the orange star the true parameters.
Size: 192.54 kB
Format: Adobe PDF
File
Access distance_mocks.pdf
Description: Examples of inference of the distance and inclination performed on simulated gravitational waves. The 50% and 90% credible intervals are derived. The blue dot represents the maximum a posteriori estimator and the orange star the true parameters.
Size: 299.07 kB
Format: Adobe PDF
File
Access chi_eff_mocks.pdf
Description: Examples of inference of the effective spin and mass ratio performed on simulated gravitational waves. The 50% and 90% credible intervals are derived. The blue dot represents the maximum a posteriori estimator and the orange star the true parameters.
Size: 235.01 kB
Format: Adobe PDF
File
Access sky_mocks.pdf
Description: Examples of inference of the sky position performed on simulated gravitational waves. The 50% and 90% credible intervals are derived. The blue dot represents the maximum a posteriori estimator and the orange star the true parameters.
Size: 271.62 kB
Format: Adobe PDF
File
Access masses_real_bayes_grid_normalized_contour_marginal_shrinked_GW150914_real_noise_corrected.pdf
Description: Inference on the masses for the GW150914 signal. MCMC stands for Markov Chain Monte-Carlo and LFI for Likelihood-Free Inference.
Size: 503.92 kB
Format: Adobe PDF
File
Access distance_real_bayes_grid_normalized_contour_marginal_GW150914_real_noise_corrected.pdf
Description: Inference on the distance and the inclination for the GW150914 signal. MCMC stands for Markov Chain Monte-Carlo and LFI for Likelihood-Free Inference.
Size: 832.77 kB
Format: Adobe PDF
File
Access effective_spin_real_bayes_grid_normalized_contour_marginal_shrinked_GW150914_real_noise_corrected.pdf
Description: Inference on the effective spin and the mass ratio for the GW150914 signal. MCMC stands for Markov Chain Monte-Carlo and LFI for Likelihood-Free Inference.
Size: 484.74 kB
Format: Adobe PDF
File
Access GW150914_normalized_skymap.pdf
Description: Inference on the sky position for the GW150914 signal. MCMC stands for Markov Chain Monte-Carlo and LFI for Likelihood-Free Inference.
Size: 64 kB
Format: Adobe PDF

Author

  • Delaunoy, Arnaud ULiège Université de Liège > Master ingé. civ. sc. don. à . fin.

Promotor(s)

Committee's member(s)

  • Geurts, Pierre ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Algorith. des syst. en interaction avec le monde physique
    ORBi View his publications on ORBi
  • Wehenkel, Louis ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Méthodes stochastiques
    ORBi View his publications on ORBi
  • Hermans, Joeri ULiège Université de Liège - ULiège >
    ORBi View his publications on ORBi
  • Total number of views 172
  • Total number of downloads 573










All documents available on MatheO are protected by copyright and subject to the usual rules for fair use.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.