Understanding deep convolutional networks
Top Cited Papers
Open Access
- 13 April 2016
- journal article
- review article
- Published by The Royal Society in Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
- Vol. 374 (2065), 20150203
- https://doi.org/10.1098/rsta.2015.0203
Abstract
Deep convolutional networks provide state-of-the-art classifications and regressions results over many high-dimensional problems. We review their architecture, which scatters data with a cascade of linear filter weights and nonlinearities. A mathematical framework is introduced to analyse their properties. Computations of invariants involve multiscale contractions with wavelets, the linearization of hierarchical symmetries and sparse separations. Applications are discussed.Keywords
Funding Information
- ERC (320959)
This publication has 19 references indexed in Scilit:
- Deep learningNature, 2015
- Intermittent process analysis with scattering momentsThe Annals of Statistics, 2015
- Deep learning of the tissue-regulated splicing codeBioinformatics, 2014
- Deep Scattering SpectrumIEEE Transactions on Signal Processing, 2014
- Invariant Scattering Convolution NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Rotation, Scaling and Deformation Invariant Scattering for Texture DiscriminationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Group Invariant ScatteringCommunications on Pure and Applied Mathematics, 2012
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Ridgelets: a key to higher-dimensional intermittency?Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 1999
- A Fast Adaptive Multipole Algorithm for Particle SimulationsSIAM Journal on Scientific and Statistical Computing, 1988