On Convergence and Generalization of Dropout Training

We study dropout in two-layer neural networks with rectified linear unit (ReLU) activations. Under mild overparametrization and assuming that the limiting kernel can separate the data distribution with a positive margin, we show that dropout training with logistic loss achieves -suboptimality in test error in O(1/ ) iterations. 34th Conference on Neural Information Processing Systems (NeurIPS 2020),