Global Minimizers of $\ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks

Re-assign community