Feedback

Faculté des Sciences appliquées
Faculté des Sciences appliquées
MASTER THESIS
VIEW 22 | DOWNLOAD 2

Master thesis : Evaluating and improving the robustness of machine learning models, using mixed-integer optimization techniques

Download
Blondiau, Sébastien ULiège
Promotor(s) : Geurts, Pierre ULiège ; Louveaux, Quentin ULiège
Date of defense : 26-Jun-2019/27-Jun-2019 • Permalink : http://hdl.handle.net/2268.2/6773
Details
Title : Master thesis : Evaluating and improving the robustness of machine learning models, using mixed-integer optimization techniques
Author : Blondiau, Sébastien ULiège
Date of defense  : 26-Jun-2019/27-Jun-2019
Advisor(s) : Geurts, Pierre ULiège
Louveaux, Quentin ULiège
Committee's member(s) : Louppe, Gilles ULiège
Wehenkel, Louis ULiège
Language : English
Number of pages : 60
Keywords : [en] robust training
[en] adversarial training
[en] adversarial examples
[en] adversarial accuracy
Discipline(s) : Engineering, computing & technology > Computer science
Target public : Researchers
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] Artificial neural networks are able to reach the highest accuracy on a great variety of complex visual tasks. Their impressive performances, often surpassing humans, attract a lot of interest.

But their opaque nature makes them considered as distrusted black-box models by experts.

In 2013, Szegedy et al. discovered that images can be slightly modified to cause the models to classify them differently. The adversary creating the modified image can even choose the new class. These modified images, called adversarial examples, draw even more distrust on these models.

In this thesis, we will present methods to evaluate the robustness of a model against such examples, among which one based on mixed integer linear programming and others based on relaxations of it.
We will also present algorithms to train models to be more robust.
Finally, we will empirically evaluate models trained with these algorithms.


File(s)

Document(s)

File
Access report.pdf
Description:
Size: 4.92 MB
Format: Adobe PDF

Annexe(s)

File
Access summary.pdf
Description:
Size: 70.12 kB
Format: Adobe PDF

Author

  • Blondiau, Sébastien ULiège Université de Liège > Master ingé. civ. info., à fin.

Promotor(s)

Committee's member(s)

  • Louppe, Gilles ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data
    ORBi View his publications on ORBi
  • Wehenkel, Louis ULiège Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Systèmes et modélisation
    ORBi View his publications on ORBi
  • Total number of views 22
  • Total number of downloads 2










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.