authors: St├ęphane Mallat
journal: Philisophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
publication year: 2016
links: arxiv (preprint), Royal Society Publishing

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.

Category: Paper Announcements