Generalized Batch Normalization: Towards Accelerating Deep Neural Networks
This work addresses the need for faster and more efficient training of deep neural networks, offering an incremental improvement over existing batch normalization methods.
The authors tackled the problem of accelerating deep neural network training by proposing a Generalized Batch Normalization (GBN) that uses alternative deviation measures and statistics, showing experimentally that it accelerates training more than conventional BN, often with improved error rates.
Utilizing recently introduced concepts from statistics and quantitative risk management, we present a general variant of Batch Normalization (BN) that offers accelerated convergence of Neural Network training compared to conventional BN. In general, we show that mean and standard deviation are not always the most appropriate choice for the centering and scaling procedure within the BN transformation, particularly if ReLU follows the normalization step. We present a Generalized Batch Normalization (GBN) transformation, which can utilize a variety of alternative deviation measures for scaling and statistics for centering, choices which naturally arise from the theory of generalized deviation measures and risk theory in general. When used in conjunction with the ReLU non-linearity, the underlying risk theory suggests natural, arguably optimal choices for the deviation measure and statistic. Utilizing the suggested deviation measure and statistic, we show experimentally that training is accelerated more so than with conventional BN, often with improved error rate as well. Overall, we propose a more flexible BN transformation supported by a complimentary theoretical framework that can potentially guide design choices.