Feedback

Faculté des Sciences appliquées
Faculté des Sciences appliquées
MASTER THESIS
VIEW 228 | DOWNLOAD 648

Master's Thesis : Differentiable Surrogate Models to Solve Nonlinear Inverse Problems

Download
Vandegar, Maxime ULiège
Promotor(s) : Louppe, Gilles ULiège
Date of defense : 25-Jun-2020/26-Jun-2020 • Permalink : http://hdl.handle.net/2268.2/9064
Details
Title : Master's Thesis : Differentiable Surrogate Models to Solve Nonlinear Inverse Problems
Author : Vandegar, Maxime ULiège
Date of defense  : 25-Jun-2020/26-Jun-2020
Advisor(s) : Louppe, Gilles ULiège
Committee's member(s) : Kagan, Michael 
Geurts, Pierre ULiège
Wehenkel, Antoine ULiège
Language : English
Number of pages : 71
Keywords : [fr] HEP, Amortized inference, Generative modeling
Discipline(s) : Engineering, computing & technology > Civil engineering
Target public : Researchers
Professionals of domain
Student
Institution(s) : Université de Liège, Liège, Belgique
Degree: Master en ingénieur civil en informatique, à finalité spécialisée en "intelligent systems"
Faculty: Master thesis of the Faculté des Sciences appliquées

Abstract

[en] Doing inference on a model defining an implicit likelihood that is not known in closed form is called likelihood-free inference. This occurs frequently in engineering and science domains where a simulator is used as a generative model of data, but the likelihood of the generated data is not known and is intractable. Given observed data, we combine the idea of hierarchical Bayesian modeling, empirical Bayes, and neural density estimation with normalizing flow to first learn a surrogate approximation of the model likelihood and then, to learn a prior distribution over the model parameters. The learned prior and the surrogate likelihood further allow to learn a posterior distribution for each observation. This is a general approach to likelihood-free inference, and is especially useful in settings where the simulator is too costly to run at inference time. We show the applicability of our methods on a real physical problem from high energy physics (HEP).


File(s)

Document(s)

File
Access MasterThesis.pdf
Description:
Size: 4.62 MB
Format: Adobe PDF

Annexe(s)

File
Access LFI_EB.zip
Description:
Size: 542.24 kB
Format: Unknown
File
Access MasterThesis_4.png
Description:
Size: 339.7 kB
Format: image/png
File
Access MasterThesis_3.png
Description:
Size: 479.87 kB
Format: image/png
File
Access MasterThesis_2.png
Description:
Size: 196.58 kB
Format: image/png
File
Access MasterThesis_1.png
Description:
Size: 218.48 kB
Format: image/png
File
Access Summary.pdf
Description:
Size: 100.13 kB
Format: Adobe PDF

Author

  • Vandegar, Maxime ULiège Université de Liège > Master ingé. civ. info., à fin.

Promotor(s)

Committee's member(s)

  • Kagan, Michael
  • 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, Antoine ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data
    ORBi View his publications on ORBi
  • Total number of views 228
  • Total number of downloads 648










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.