Convolutional Kernel Networks

Julien Mairal 1, * Piotr Koniusz 1, * Zaid Harchaoui 1, * Cordelia Schmid 1, *
* Auteur correspondant
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : An important goal in visual recognition is to devise image representations that are invariant to particular transformations. In this paper, we address this goal with a new type of convolutional neural network (CNN) whose invariance is encoded by a reproducing kernel. Unlike traditional approaches where neural networks are learned either to represent data or for solving a classification task, our network learns to approximate the kernel feature map on training data. Such an approach enjoys several benefits over classical ones. First, by teaching CNNs to be invariant, we obtain simple network architectures that achieve a similar accuracy to more complex ones, while being easy to train and robust to overfitting. Second, we bridge a gap between the neural network literature and kernels, which are natural tools to model invariance. We evaluate our methodology on visual recognition tasks where CNNs have proven to perform well, e.g., digit recognition with the MNIST dataset, and the more challenging CIFAR-10 and STL-10 datasets, where our accuracy is competitive with the state of the art.
Type de document :
Communication dans un congrès
Advances in Neural Information Processing Systems (NIPS), Dec 2014, Montreal, Canada. <http://nips.cc>
Liste complète des métadonnées


https://hal.inria.fr/hal-01005489
Contributeur : Julien Mairal <>
Soumis le : mercredi 12 novembre 2014 - 09:43:50
Dernière modification le : mardi 11 août 2015 - 01:05:12
Document(s) archivé(s) le : vendredi 13 février 2015 - 10:26:01

Fichiers

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

Identifiants

  • HAL Id : hal-01005489, version 3
  • ARXIV : 1406.3332

Collections

Citation

Julien Mairal, Piotr Koniusz, Zaid Harchaoui, Cordelia Schmid. Convolutional Kernel Networks. Advances in Neural Information Processing Systems (NIPS), Dec 2014, Montreal, Canada. <http://nips.cc>. <hal-01005489v3>

Partager

Métriques

Consultations de
la notice

1267

Téléchargements du document

743