Stabilize deep ResNet with a sharp scaling factor tau

doi:10.1007/s10994-022-06192-x

	Stabilize deep ResNet with a sharp scaling factor tau
	Zhang, Huishuai 1; Yu, Da 2; Yi, Mingyang 3; Chen, Wei 1,4; Liu, Tie-Yan
	2022-08-01
发表期刊	MACHINE LEARNING
ISSN	0885-6125
页码	34
摘要	We study the stability and convergence of training deep ResNets with gradient descent. Specifically, we show that the parametric branch in the residual block should be scaled down by a factor tau=O(1/L) to guarantee stable forward/backward process, where L is the number of residual blocks. Moreover, we establish a converse result that the forward process is unbounded when tau > L-1/2+c for any positive constant c. The above two results together establish a sharp value of the scaling factor in determining the stability of deep ResNet. Based on the stability result, we further show that gradient descent finds the global minima if the ResNet is properly over-parameterized, which significantly improves over the previous work with a much larger range of tau that admits global convergence. Moreover, we show that the convergence rate is independent of the depth, theoretically justifying the advantage of ResNet over vanilla feedforward network. Empirically, with such a factor tau, one can train deep ResNet without normalization layer. Moreover for ResNets with normalization layer, adding such a factor tau also stabilizes the training and obtains significant performance gain for deep ResNet.
DOI	10.1007/s10994-022-06192-x
收录类别	SCI
语种	英语
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000834697900001
出版者	SPRINGER
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/19483
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zhang, Huishuai; Chen, Wei
作者单位	1.Microsoft Res Asia, Beijing, Peoples R China 2.Sun Yat Sen Univ, Guangzhou, Peoples R China 3.Univ Chinese Acad Sci, Beijing, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Huishuai,Yu, Da,Yi, Mingyang,et al. Stabilize deep ResNet with a sharp scaling factor tau[J]. MACHINE LEARNING,2022:34.
APA	Zhang, Huishuai,Yu, Da,Yi, Mingyang,Chen, Wei,&Liu, Tie-Yan.(2022).Stabilize deep ResNet with a sharp scaling factor tau.MACHINE LEARNING,34.
MLA	Zhang, Huishuai,et al."Stabilize deep ResNet with a sharp scaling factor tau".MACHINE LEARNING (2022):34.