Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift


Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs. Our method draws its strength from making nor-malization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization, and in some cases eliminates the need for Dropout. Applied to a state-of-the-art image classification model, Batch Nor-malization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin. Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.82% top-5 test error, exceeding the accuracy of human raters.

Extracted Key Phrases

Showing 1-3 of 3 references

Batch Normalization Presentation

  • Sergey Ioffe
  • 2015
1 Excerpt
Showing 1-10 of 1,510 extracted citations
Citations per Year

2,475 Citations

Semantic Scholar estimates that this publication has received between 2,275 and 2,695 citations based on the available data.

See our FAQ for additional information.