Easy Batch Normalization
This addresses a novel aspect of training data utilization for neural networks, but it appears incremental as it builds on existing batch normalization techniques.
The paper tackles the problem of improving neural network training by exploring the use of easy examples, proposing an auxiliary batch normalization method to enhance both standard and robust accuracy.
It was shown that adversarial examples improve object recognition. But what about their opposite side, easy examples? Easy examples are samples that the machine learning model classifies correctly with high confidence. In our paper, we are making the first step toward exploring the potential benefits of using easy examples in the training procedure of neural networks. We propose to use an auxiliary batch normalization for easy examples for the standard and robust accuracy improvement.