For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Symmetric Decomposition of Convolution Kernels
Jun OU Yujian LI
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2019/01/01
Online ISSN: 1745-1361
Type of Manuscript: LETTER
Category: Biocybernetics, Neurocomputing
symmetric decomposition, convolution kernels, speedup, reduction of network parameters,
Full Text: PDF(1.1MB)
>>Buy this Article
It is a hot issue that speeding up the network layers and decreasing the network parameters in convolutional neural networks (CNNs). In this paper, we propose a novel method, namely, symmetric decomposition of convolution kernels (SDKs). It symmetrically separates k×k convolution kernels into (k×1 and 1×k) or (1×k and k×1) kernels. We conduct the comparison experiments of the network models designed by SDKs on MNIST and CIFAR-10 datasets. Compared with the corresponding CNNs, we obtain good recognition performance, with 1.1×-1.5× speedup and more than 30% reduction of network parameters. The experimental results indicate our method is useful and effective for CNNs in practice, in terms of speedup performance and reduction of parameters.