Accelerating Training of Deep Neural Networks with a Standardization Loss

Re-assign community