Decorrelated Batch Normalization

TL;DR

Decorrelated Batch Normalization (DBN) extends standard BN by whitening activations, improving training efficiency and accuracy in deep neural networks, especially residual networks, through effective whitening techniques like ZCA.

Contribution

This paper introduces DBN, a whitening-based normalization method that overcomes issues with PCA whitening and enhances deep network training and generalization.

Findings

DBN improves training speed and accuracy over BN.

ZCA whitening in DBN avoids stochastic axis swapping.

DBN enhances residual network performance on CIFAR and ImageNet.

Abstract

Batch Normalization (BN) is capable of accelerating the training of deep models by centering and scaling activations within mini-batches. In this work, we propose Decorrelated Batch Normalization (DBN), which not just centers and scales activations but whitens them. We explore multiple whitening techniques, and find that PCA whitening causes a problem we call stochastic axis swapping, which is detrimental to learning. We show that ZCA whitening does not suffer from this problem, permitting successful learning. DBN retains the desirable qualities of BN and further improves BN's optimization efficiency and generalization ability. We design comprehensive experiments to show that DBN can improve the performance of BN on multilayer perceptrons and convolutional neural networks. Furthermore, we consistently improve the accuracy of residual networks on CIFAR-10, CIFAR-100, and ImageNet.