Learning Sparse Neural Networks via ℓ0 and Tℓ1 by a Relaxed Variable Splitting Method with Application to Multi-scale Curve Classification
Learning Sparse Neural Networks via ℓ0 and Tℓ1 by a Relaxed Variable Splitting Method with Application to Multi-scale Curve Classification
复制标题
通过松弛变量分裂方法通过α0和Tα1学习稀疏神经网络并应用于多尺度曲线分类
DOI：
10.1007/978-3-030-21803-4
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Xin, J
中科院分区：
文献类型：
--
作者：
Xue, F;Xin, J
We study sparsification of convolutional neural networks (CNN) by a relaxed variable splitting method ofand transformed-(T) penalties, with application to complex curves such as texts written in different fonts, and words written with trembling hands simulating those of Parkinson’s disease patients. The CNN contains 3 convolutional layers, each followed by a maximum pooling, and finally a fully connected layer which contains the largest number of network weights. Withpenalty, we achieved over 99% test accuracy in distinguishing shaky vs. regular fonts or hand writings with above 86% of the weights in the fully connected layer being zero. Comparable sparsity and test accuracy are also reached with a proper choice of Tpenalty.
