Improving Deep Neural Networks by Using Sparse Dropout Strategy

	Improving Deep Neural Networks by Using Sparse Dropout Strategy
	Zheng Hao; Mingming Chen; Wenju Liu; Zhanlei Yang; Shan Liang; Hao Zheng
	2014
会议名称	ChinaSIP
会议录名称	ChinaSIP
会议日期	2014
会议地点	Xi an, Shanxi, China
摘要	Recently, deep neural networks(DNNs) have achieved excelleng results on benchmarks for acoustic modeling of speech recognition. By randomly discarding network units, a strategy which is called as dropout can improve the performance of DNNs by reducing the influence of over-fitting. However, the random dropout strategy treats units indiscriminately, which may lose information on distributions of units outputs. In this paper, we improve the dropout strategy by differential treatment to units according to their outputs. Only minor changes to an existing neural network system can achieve a significant improvement. Experiments of phone recognition on TIMIT show that the sparse dropout fine-tuning gets significant performance improvement.; Recently, deep neural networks(DNNs) have achieved excelleng results on benchmarks for acoustic modeling of speech recognition. By randomly discarding network units, a strategy which is called as dropout can improve the performance of DNNs by reducing the influence of over-fitting. However, the random dropout strategy treats units indiscriminately, which may lose information on distributions of units outputs. In this paper, we improve the dropout strategy by differential treatment to units according to their outputs. Only minor changes to an existing neural network system can achieve a significant improvement. Experiments of phone recognition on TIMIT show that the sparse dropout fine-tuning gets significant performance improvement.
关键词	Dropout Sparse Dropout Deep Neural Networks Deep Learning
收录类别	EI
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/11776
专题	多模态人工智能系统全国重点实验室_机器人视觉
通讯作者	Hao Zheng
作者单位	National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences
推荐引用方式 GB/T 7714	Zheng Hao,Mingming Chen,Wenju Liu,et al. Improving Deep Neural Networks by Using Sparse Dropout Strategy[C],2014.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
ChinaSIP-2014-1.pdf（146KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载