Deep neural networks approximate smooth functions more efficiently than theory suggested

Original: Approximation and learning of anisotropic and mixed smooth functions by deep ReLU neural networks

Source: arxiv.org ↗