Mitigating the Curse of Dimensionality in Uniform Convergence of Deep Neural Networks via Smooth Activations

Open in new window