Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

Yuanzhi Li, Colin Wei, Tengyu Ma. Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks. In Hanna M. Wallach, Hugo Larochelle, Alina Beygelzimer, Florence d'Alché-Buc, Edward A. Fox, Roman Garnett, editors, Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, 8-14 December 2019, Vancouver, BC, Canada. pages 11669-11680, 2019. [doi]

Abstract

Abstract is missing.