Equi-normalization of Neural Networks
2019-02-27ICLR 2019Code Available0· sign in to hype
Pierre Stock, Benjamin Graham, Rémi Gribonval, Hervé Jégou
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/facebookresearch/enormOfficialIn paperpytorch★ 0
Abstract
Modern neural networks are over-parametrized. In particular, each rectified linear hidden unit can be modified by a multiplicative factor by adjusting input and output weights, without changing the rest of the network. Inspired by the Sinkhorn-Knopp algorithm, we introduce a fast iterative method for minimizing the L2 norm of the weights, equivalently the weight decay regularizer. It provably converges to a unique solution. Interleaving our algorithm with SGD during training improves the test accuracy. For small batches, our approach offers an alternative to batch-and group-normalization on CIFAR-10 and ImageNet with a ResNet-18.