Provable Generalization Bounds for Deep Neural Networks with Momentum-Adaptive Gradient Dropout

Open in new window