Understanding Deep Convolutional Networks
Machine Learning
2016-04-27 v1 Computer Vision and Pattern Recognition
Machine Learning
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 non-linearities. A mathematical framework is introduced to analyze their properties. Computations of invariants involve multiscale contractions, the linearization of hierarchical symmetries, and sparse separations. Applications are discussed.
Cite
@article{arxiv.1601.04920,
title = {Understanding Deep Convolutional Networks},
author = {Stéphane Mallat},
journal= {arXiv preprint arXiv:1601.04920},
year = {2016}
}
Comments
17 pages, 4 Figures