Convolutional Neural Fabrics

Shreyas Saxena 1 Jakob Verbeek 1
1 Thoth - Apprentissage de modèles à partir de données massives
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann
Abstract : Despite the success of CNNs, selecting the optimal architecture for a given task remains an open problem. Instead of aiming to select a single optimal architecture, we propose a " fabric " that embeds an exponentially large number of architectures. The fabric consists of a 3D trellis that connects response maps at different layers, scales, and channels with a sparse homogeneous local connectivity pattern. The only hyper-parameters of a fabric are the number of channels and layers. While individual architectures can be recovered as paths, the fabric can in addition ensemble all embedded architectures together, sharing their weights where their paths overlap. Parameters can be learned using standard methods based on back-propagation, at a cost that scales linearly in the fabric size. We present benchmark results competitive with the state of the art for image classification on MNIST and CIFAR10, and for semantic segmentation on the Part Labels dataset.
Type de document :
Communication dans un congrès
Advances in Neural Information Processing Systems (NIPS), Dec 2016, Barcelona, Spain
Liste complète des métadonnées


https://hal.inria.fr/hal-01359150
Contributeur : Thoth Team <>
Soumis le : lundi 30 janvier 2017 - 13:41:46
Dernière modification le : vendredi 3 mars 2017 - 23:43:52

Fichier

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

Identifiants

  • HAL Id : hal-01359150, version 3

Collections

Citation

Shreyas Saxena, Jakob Verbeek. Convolutional Neural Fabrics. Advances in Neural Information Processing Systems (NIPS), Dec 2016, Barcelona, Spain. <hal-01359150v3>

Partager

Métriques

Consultations de
la notice

515

Téléchargements du document

204