Leverage Score Sampling for Complete Mode ...
Type de document :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Titre :
Leverage Score Sampling for Complete Mode Coverage in Generative Adversarial Networks
Auteur(s) :
Schreurs, Joachim [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
de Meulemeester, Hannes [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Fanuel, Michaël [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Centre National de la Recherche Scientifique [CNRS]
de Moor, Bart [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Suykens, Johan A. K. [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
de Meulemeester, Hannes [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Fanuel, Michaël [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Centre National de la Recherche Scientifique [CNRS]
de Moor, Bart [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Suykens, Johan A. K. [Auteur]
Department of Electrical Engineering [KU Leuven] [KU-ESAT]
Titre de la manifestation scientifique :
The 7th International Online & Onsite Conference on Machine Learning, Optimization, and Data Science
Ville :
Grasmere
Pays :
Royaume-Uni
Date de début de la manifestation scientifique :
2021-10-04
Discipline(s) HAL :
Sciences de l'ingénieur [physics]
Mathématiques [math]
Informatique [cs]
Mathématiques [math]
Informatique [cs]
Résumé en anglais : [en]
Commonly, machine learning models minimize an empirical expectation. As a result, the trained models typically perform well for the majority of the data but the performance may deteriorate in less dense regions of the ...
Lire la suite >Commonly, machine learning models minimize an empirical expectation. As a result, the trained models typically perform well for the majority of the data but the performance may deteriorate in less dense regions of the dataset. This issue also arises in generative modeling. A generative model may overlook underrepresented modes that are less frequent in the empirical data distribution. This problem is known as complete mode coverage. We propose a sampling procedure based on ridge leverage scores which significantly improves mode coverage when compared to standard methods and can easily be combined with any GAN. Ridge leverage scores are computed by using an explicit feature map, associated with the next-to-last layer of a GAN discriminator or of a pre-trained network, or by using an implicit feature map corresponding to a Gaussian kernel. Multiple evaluations against recent approaches of complete mode coverage show a clear improvement when using the proposed sampling strategy.Lire moins >
Lire la suite >Commonly, machine learning models minimize an empirical expectation. As a result, the trained models typically perform well for the majority of the data but the performance may deteriorate in less dense regions of the dataset. This issue also arises in generative modeling. A generative model may overlook underrepresented modes that are less frequent in the empirical data distribution. This problem is known as complete mode coverage. We propose a sampling procedure based on ridge leverage scores which significantly improves mode coverage when compared to standard methods and can easily be combined with any GAN. Ridge leverage scores are computed by using an explicit feature map, associated with the next-to-last layer of a GAN discriminator or of a pre-trained network, or by using an implicit feature map corresponding to a Gaussian kernel. Multiple evaluations against recent approaches of complete mode coverage show a clear improvement when using the proposed sampling strategy.Lire moins >
Langue :
Anglais
Comité de lecture :
Oui
Audience :
Internationale
Vulgarisation :
Non
Collections :
Source :
Fichiers
- http://arxiv.org/pdf/2104.02373
- Accès libre
- Accéder au document
- 2104.02373
- Accès libre
- Accéder au document