Titre : |
Optimal transport theory for anomaly detection |
Type de document : |
document électronique |
Auteurs : |
Imad Anis Kheffache, Auteur ; Mireille Batton-hubert, Directeur de thèse ; Marina Krémé, Directeur de thèse ; Arthur Kramer, Directeur de thèse ; Latifa Debbi, Directeur de thèse |
Editeur : |
[S.l.] : [s.n.] |
Année de publication : |
2024 |
Importance : |
1 fichier PDF (2.1 Mo) |
Présentation : |
ill. |
Note générale : |
Mode d'accès : accès au texte intégral par intranet.
Mémoire de Projet de Fin d’Etudes : Génie Industriel. Data Science-Intelligence Artificielle : Alger, Ecole Nationale Polytechnique : 2024
Bibliogr. p. 60 - 63 |
Langues : |
Anglais (eng) |
Mots-clés : |
Optimal transport
Cost matrix
Anomalies
Anomaly detection
Signals
Threshold |
Index. décimale : |
PI02824 |
Résumé : |
The emergence of Kantorovich’s formulation for the optimal transport problem and the advent of powerful computational resources have made significant advancements possible in several fields such as economics, fluid dynamics, and computer vision. Notably, it has also enabled improvements in various branches of data science, including computer vision and machine learning, among others. The goal of this thesis was to explore the link between optimal transport and anomaly detection and to address the fol- lowing research question: Is it possible to leverage optimal transport theory and its tools for anomaly detection?
In this work, we provide an affirmative answer to the guiding question. We also present preliminary methods that use optimal transport to detect anomalies. |
Optimal transport theory for anomaly detection [document électronique] / Imad Anis Kheffache, Auteur ; Mireille Batton-hubert, Directeur de thèse ; Marina Krémé, Directeur de thèse ; Arthur Kramer, Directeur de thèse ; Latifa Debbi, Directeur de thèse . - [S.l.] : [s.n.], 2024 . - 1 fichier PDF (2.1 Mo) : ill. Mode d'accès : accès au texte intégral par intranet.
Mémoire de Projet de Fin d’Etudes : Génie Industriel. Data Science-Intelligence Artificielle : Alger, Ecole Nationale Polytechnique : 2024
Bibliogr. p. 60 - 63 Langues : Anglais ( eng)
Mots-clés : |
Optimal transport
Cost matrix
Anomalies
Anomaly detection
Signals
Threshold |
Index. décimale : |
PI02824 |
Résumé : |
The emergence of Kantorovich’s formulation for the optimal transport problem and the advent of powerful computational resources have made significant advancements possible in several fields such as economics, fluid dynamics, and computer vision. Notably, it has also enabled improvements in various branches of data science, including computer vision and machine learning, among others. The goal of this thesis was to explore the link between optimal transport and anomaly detection and to address the fol- lowing research question: Is it possible to leverage optimal transport theory and its tools for anomaly detection?
In this work, we provide an affirmative answer to the guiding question. We also present preliminary methods that use optimal transport to detect anomalies. |
|