A Mathematical Theory of Deep Convolutional Neural Networks for Feature Extraction

DOI10.1109/TIT.2017.2776228zbMath1390.94053arXiv1512.06293OpenAlexW2962957157MaRDI QIDQ4566749

Helmut Bölcskei, Thomas Wiatowski

Publication date: 27 June 2018

Published in: IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1512.06293

Mathematics Subject Classification ID

Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Related Items

Boosting Poisson regression models with telematics car driving data, Asymptotic analysis of higher-order scattering transform of Gaussian processes, What can we learn from telematics car driving data: a survey, Prevalence of neural collapse during the terminal phase of deep learning training, A multi-hierarchical attention-based prediction method on time series with spatio-temporal context among variables, Optimal approximation of piecewise smooth functions using deep ReLU neural networks, Deep limits of residual neural networks, Approximation Analysis of Convolutional Neural Networks, Central and Noncentral Limit Theorems Arising from the Scattering Transform and Its Neural Activation Generalization, Geometric scattering on measure spaces, Analysis of time-frequency scattering transforms, Identifiability of latent-variable and structural-equation models: from linear to nonlinear, Stability of the scattering transform for deformations with minimal regularity, Stability of iterated dyadic filter banks, On generalizations of the nonwindowed scattering transform, Affine symmetries and neural network identifiability, Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review, Lipschitz properties for deep convolutional networks, Deep Convolutional Framelets: A General Deep Learning Framework for Inverse Problems, Rotationally invariant time-frequency scattering transforms, Time-series forecasting of mortality rates using deep learning, Gabor neural networks with proven approximation properties, Provably scale-covariant continuous hierarchical networks based on scale-normalized differential expressions coupled in cascade, Some open questions on morphological operators and representations in the deep learning era. A personal vision, Convolutional spectral kernel learning with generalization guarantees, A dilated convolution network-based LSTM model for multi-step prediction of chaotic time series, Unnamed Item, Approximating functions with multi-features by deep convolutional neural networks, Neural network identifiability for a family of sigmoidal nonlinearities