DeepAI AI Chat
Log In Sign Up

Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks

by   Bum Jun Kim, et al.
POSTECH / 포항공과대학교

L2 regularization for weights in neural networks is widely used as a standard training trick. However, L2 regularization for gamma, a trainable parameter of batch normalization, remains an undiscussed mystery and is applied in different ways depending on the library and practitioner. In this paper, we study whether L2 regularization for gamma is valid. To explore this issue, we consider two approaches: 1) variance control to make the residual network behave like identity mapping and 2) stable optimization through the improvement of effective learning rate. Through two analyses, we specify the desirable and undesirable gamma to apply L2 regularization and propose four guidelines for managing them. In several experiments, we observed the increase and decrease in performance caused by applying L2 regularization to gamma of four categories, which is consistent with our four guidelines. Our proposed guidelines were validated through various tasks and architectures, including variants of residual networks and transformers.


page 1

page 2

page 3

page 4


L2 Regularization versus Batch and Weight Normalization

Batch Normalization is a commonly used trick to improve the training of ...

How to Use Dropout Correctly on Residual Networks with Batch Normalization

For the stable optimization of deep neural networks, regularization meth...

Batch Normalization Biases Deep Residual Networks Towards Shallow Paths

Batch normalization has multiple benefits. It improves the conditioning ...

Fixup Initialization: Residual Learning Without Normalization

Normalization layers are a staple in state-of-the-art deep neural networ...

Theroretical Insight into Batch Normalization: Data Dependant Auto-Tuning of Regularization Rate

Batch normalization is widely used in deep learning to normalize interme...

On Layer Normalization in the Transformer Architecture

The Transformer is widely used in natural language processing tasks. To ...

Spherical Motion Dynamics of Deep Neural Networks with Batch Normalization and Weight Decay

We comprehensively reveal the learning dynamics of deep neural networks ...