Add papers and vote!
How to: one “answer” = one paper ; vote and comment on proposed papers.
Valentin Emiya Unselected an answer
Some theoretical insights into the benefits of deep learning
Chizat, L., & Bach, F. Implicit bias of gradient descent for wide two-layer neural networks trained with the logistic loss. (COLT 2020) [pdf]
Thomas Schatz Edited answer