Regularization in ResNet with Stochastic Depth

Abstract

Regularization plays a major role in modern deep learning. From classic techniques such as L1,L2 penalties to other noise-based methods such as Dropout, regularization often yields better generalization properties by avoiding overfitting. Recently, Stochastic Depth (SD) has emerged as an alternative regularization technique for residual neural networks (ResNets) and has proven to boost the performance of ResNet on many tasks [Huang et al., 2016]. Despite the recent success of SD, little is known about this technique from a theoretical perspective. This paper provides a hybrid analysis combining perturbation analysis and signal propagation to shed light on different regularization effects of SD. Our analysis allows us to derive principled guidelines for choosing the survival rates used for training with SD.

24 pages, 15 figures

References in corpus (5)

• Improving neural networks by preventing co-adaptation of feature detectors
• On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima
• On the Impact of the Activation Function on Deep Neural Networks Training
• Dropout as a Low-Rank Regularizer for Matrix Factorization
• On Dropout and Nuclear Norm Regularization