Follow
Global minimizers, strict and non-strict saddle points, and implicit regularization for deep linear neural networks. (arXiv:2107.13289v1 [math.ST]) http://arxiv.org/abs/2107.13289
Global minimizers, strict and non-strict saddle points, and implicit regularization for deep linear neural networks. (arXiv:2107.13289v1 [math.ST]) http://arxiv.org/abs/2107.13289