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

Abstract

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 normalization 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. It also acts as a regularizer, in some cases eliminating the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization 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 4.9% top-5 validation error (and 4.8% test error), exceeding the accuracy of human raters.

References in corpus (4)

• On the difficulty of training Recurrent Neural Networks
• Going Deeper with Convolutions
• Deep Image: Scaling up Image Recognition
• Natural Neural Networks

Cited by in corpus (16)

• Quantized Neural Networks: Training Neural Networks with Low Precision Weights and Activations
• Understanding deep learning requires rethinking generalization
• FINN: A Framework for Fast, Scalable Binarized Neural Network Inference
• Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model
• Show and Tell: Lessons learned from the 2015 MSCOCO Image Captioning Challenge
• Residual Networks of Residual Networks: Multilevel Residual Networks
• Deep Image: Scaling up Image Recognition
• Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approach
• A Fully Convolutional Deep Auditory Model for Musical Chord Recognition
• Deep Learning with Low Precision by Half-wave Gaussian Quantization
• Deep learning for plasma tomography using the bolometer system at JET
• Crafting GBD-Net for Object Detection
• DeepWarp: Photorealistic Image Resynthesis for Gaze Manipulation
• Byte-based Language Identification with Deep Convolutional Networks
• A dataset and exploration of models for understanding video data through fill-in-the-blank question-answering
• Learning Identity Mappings with Residual Gates