On the Heavy-Tailed Theory of Stochastic Gradient Descent for Deep Neural Networks. (arXiv:1912.00018v1 [stat.ML]) http://arxiv.org/abs/1912.00018
CleverLibre Social is an inclusive social instance for open discussion, learning, and community. All cultures welcome. Hate speech and harassment strictly forbidden.