Automatic Node Selection for Deep Neural Networks using Group Lasso Regularization
Computation and Language
2016-11-18 v1 Machine Learning
Machine Learning
Abstract
We examine the effect of the Group Lasso (gLasso) regularizer in selecting the salient nodes of Deep Neural Network (DNN) hidden layers by applying a DNN-HMM hybrid speech recognizer to TED Talks speech data. We test two types of gLasso regularization, one for outgoing weight vectors and another for incoming weight vectors, as well as two sizes of DNNs: 2048 hidden layer nodes and 4096 nodes. Furthermore, we compare gLasso and L2 regularizers. Our experiment results demonstrate that our DNN training, in which the gLasso regularizer was embedded, successfully selected the hidden layer nodes that are necessary and sufficient for achieving high classification power.
Cite
@article{arxiv.1611.05527,
title = {Automatic Node Selection for Deep Neural Networks using Group Lasso Regularization},
author = {Tsubasa Ochiai and Shigeki Matsuda and Hideyuki Watanabe and Shigeru Katagiri},
journal= {arXiv preprint arXiv:1611.05527},
year = {2016}
}
Comments
Submitted to ICASSP 2017