Hongyang Zhang, Junru Shao, Ruslan Salakhutdinov. Deep Neural Networks with Multi-Branch Architectures Are Intrinsically Less Non-Convex. In Kamalika Chaudhuri, Masashi Sugiyama, editors, The 22nd International Conference on Artificial Intelligence and Statistics, AISTATS 2019, 16-18 April 2019, Naha, Okinawa, Japan. Volume 89 of Proceedings of Machine Learning Research, pages 1099-1109, PMLR, 2019. [doi]
Abstract is missing.