Convergence of SGD for Training Neural Networks with Sliced Wasserstein Losses

Re-assign community