i-RevNet: Deep Invertible Networks - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

i-RevNet: Deep Invertible Networks

Résumé

It is widely believed that the success of deep convolutional networks is based on progressively discarding uninformative variability about the input with respect to the problem at hand. This is supported empirically by the difficulty of recovering images from their hidden representations, in most commonly used network architectures. In this paper we show via a one-to-one mapping that this loss of information is not a necessary condition to learn representations that generalize well on complicated problems, such as ImageNet. Via a cascade of homeomorphic layers, we build the i-RevNet, a network that can be fully inverted up to the final projection onto the classes, i.e. no information is discarded. Building an invertible architecture is difficult, for one, because the local inversion is ill-conditioned, we overcome this by providing an explicit inverse. An analysis of i-RevNets learned representations suggests an alternative explanation for the success of deep networks by a progressive contraction and linear separation with depth. To shed light on the nature of the model learned by the i-RevNet we reconstruct linear interpolations between natural image representations.
Fichier principal
Vignette du fichier
iclr2018_conference_final_submitted.pdf (1.78 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01712808 , version 1 (20-02-2018)

Identifiants

Citer

Jörn-Henrik Jacobsen, Arnold Smeulders, Edouard Oyallon. i-RevNet: Deep Invertible Networks. ICLR 2018 - International Conference on Learning Representations, Apr 2018, Vancouver, Canada. ⟨hal-01712808⟩
1046 Consultations
1819 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More