Towards Good Practice in Large-Scale Learning for Image Classification

Florent Perronnin 1 Zeynep Akata 1, 2 Zaid Harchaoui 2 Cordelia Schmid 2
2 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : We propose a benchmark of several objective functions for large-scale image classification: we compare the one- vs-rest, multiclass, ranking and weighted average ranking SVMs. Using stochastic gradient descent optimization, we can scale the learning to millions of images and thousands of classes. Our experimental evaluation shows that ranking based algorithms do not outperform a one-vs-rest strategy and that the gap between the different algorithms reduces in case of high-dimensional data. We also show that for one-vs-rest, learning through cross-validation the optimal degree of imbalance between the positive and the negative samples can have a significant impact. Furthermore, early stopping can be used as an effective regularization strategy when training with stochastic gradient algorithms. Follow- ing these "good practices", we were able to improve the state-of-the-art on a large subset of 10K classes and 9M of images of ImageNet from 16.7% accuracy to 19.1%.
Type de document :
Communication dans un congrès
CVPR 2012 - IEEE Computer Vision and Pattern Recognition, Jun 2012, Providence (RI), United States. IEEE, pp.3482-3489, 2012, 〈10.1109/CVPR.2012.6248090〉
Liste complète des métadonnées

Littérature citée [38 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-00690014
Contributeur : Thoth Team <>
Soumis le : vendredi 20 avril 2012 - 18:39:35
Dernière modification le : mercredi 11 avril 2018 - 01:59:40
Document(s) archivé(s) le : samedi 21 juillet 2012 - 02:35:37

Fichier

cvpr2012.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

Citation

Florent Perronnin, Zeynep Akata, Zaid Harchaoui, Cordelia Schmid. Towards Good Practice in Large-Scale Learning for Image Classification. CVPR 2012 - IEEE Computer Vision and Pattern Recognition, Jun 2012, Providence (RI), United States. IEEE, pp.3482-3489, 2012, 〈10.1109/CVPR.2012.6248090〉. 〈hal-00690014〉

Partager

Métriques

Consultations de la notice

3548

Téléchargements de fichiers

5172