Batch Normalization_ Accelerating Deep Network Training b.pdf
Binarized Neural Networks_ Training Neural Networks with Weights and Activations Constrained to+ 1 or−1
Decoupled Neural Interfaces using Synthetic Gradients
Dropout_ A Simple Way to Prevent Neural Networks from
Improving neural networks by preventing
Layer Normalization
Net2net_ Accelerating learning via knowledge transfer
Network Morphism
