Skip to Main content Skip to Navigation
New interface
Conference papers

Deep Network Classification by Scattering and Homotopy Dictionary Learning

Abstract : We introduce a sparse scattering deep convolutional neural network, which provides a simple model to analyze properties of deep representation learning for classification. Learning a single dictionary matrix with a classifier yields a higher classification accuracy than AlexNet over the ImageNet 2012 dataset. The network first applies a scattering transform that linearizes variabilities due to geometric transformations such as translations and small deformations. A sparse l1 dictionary coding reduces intra-class variability while preserving class separation through projections over unions of linear spaces. It is implemented in a deep convolutional network with a homotopy algorithm having an exponential convergence. A convergence proof is given in a general framework that includes ALISTA. Classification results are analyzed on ImageNet.
Complete list of metadata

Cited literature [31 references]  Display  Hide  Download
Contributor : John Zarka Connect in order to contact the contributor
Submitted on : Friday, October 23, 2020 - 4:20:09 PM
Last modification on : Thursday, March 17, 2022 - 10:08:39 AM
Long-term archiving on: : Sunday, January 24, 2021 - 7:16:49 PM


Files produced by the author(s)


  • HAL Id : hal-02976813, version 1



John Zarka, Louis Thiry, Tomás Angles, Stephane Mallat. Deep Network Classification by Scattering and Homotopy Dictionary Learning. ICLR 2020 - 8th International Conference on Learning Representations, Apr 2020, Addis Ababa / Virtual, Ethiopia. ⟨hal-02976813⟩



Record views


Files downloads