Scaling the Scattering Transform: Deep Hybrid Networks - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2017

Scaling the Scattering Transform: Deep Hybrid Networks

Résumé

We use the scattering network as a generic and fixed ini-tialization of the first layers of a supervised hybrid deep network. We show that early layers do not necessarily need to be learned, providing the best results to-date with pre-defined representations while being competitive with Deep CNNs. Using a shallow cascade of 1 × 1 convo-lutions, which encodes scattering coefficients that correspond to spatial windows of very small sizes, permits to obtain AlexNet accuracy on the imagenet ILSVRC2012. We demonstrate that this local encoding explicitly learns in-variance w.r.t. rotations. Combining scattering networks with a modern ResNet, we achieve a single-crop top 5 error of 11.4% on imagenet ILSVRC2012, comparable to the Resnet-18 architecture, while utilizing only 10 layers. We also find that hybrid architectures can yield excellent performance in the small sample regime, exceeding their end-to-end counterparts, through their ability to incorporate geometrical priors. We demonstrate this on subsets of the CIFAR-10 dataset and by setting a new state-of-the-art on the STL-10 dataset.
Fichier principal
Vignette du fichier
iccv2017_conference.pdf (218.31 Ko) Télécharger le fichier
bullsit-eps-converted-to.pdf (9.04 Ko) Télécharger le fichier
little_wood-eps-converted-to.pdf (12.41 Ko) Télécharger le fichier
little_wood_1d-eps-converted-to.pdf (12.81 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01495734 , version 1 (27-03-2017)
hal-01495734 , version 2 (03-04-2017)

Licence

Licence Ouverte - etalab

Identifiants

Citer

Edouard Oyallon, Eugene Belilovsky, Sergey Zagoruyko. Scaling the Scattering Transform: Deep Hybrid Networks. 2017. ⟨hal-01495734v1⟩
867 Consultations
936 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More