Back to Search
Start Over
Universal Consistency of Deep Convolutional Neural Networks.
- Source :
-
IEEE Transactions on Information Theory . Jul2022, Vol. 68 Issue 7, p4610-4617. 8p. - Publication Year :
- 2022
-
Abstract
- Compared with avid research activities of deep convolutional neural networks (DCNNs) in practice, the study of theoretical behaviors of DCNNs lags heavily behind. In particular, the universal consistency of DCNNs remains open. In this paper, we prove that implementing empirical risk minimization on DCNNs with expansive convolution (with zero-padding) is strongly universally consistent. Motivated by the universal consistency, we conduct a series of experiments to show that without any fully connected layers, DCNNs with expansive convolution perform not worse than the widely used deep neural networks with hybrid structure containing contracting (without zero-padding) convolutional layers and several fully connected layers. [ABSTRACT FROM AUTHOR]
- Subjects :
- *CONVOLUTIONAL neural networks
*DEEP learning
Subjects
Details
- Language :
- English
- ISSN :
- 00189448
- Volume :
- 68
- Issue :
- 7
- Database :
- Academic Search Index
- Journal :
- IEEE Transactions on Information Theory
- Publication Type :
- Academic Journal
- Accession number :
- 157551867
- Full Text :
- https://doi.org/10.1109/TIT.2022.3151753