Batch Normalization Is Blind to the First and Second Derivatives of the Loss

Zhanpeng Zhou, Wen Shen 0002, Huixin Chen, Ling Tang, Yuefeng Chen, Quanshi Zhang. Batch Normalization Is Blind to the First and Second Derivatives of the Loss. In Michael J. Wooldridge, Jennifer G. Dy, Sriraam Natarajan, editors, Thirty-Eigth AAAI Conference on Artificial Intelligence, AAAI 2024, Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence, IAAI 2024, Fourteenth Symposium on Educational Advances in Artificial Intelligence, EAAI 2014, February 20-27, 2024, Vancouver, Canada. pages 20010-20018, AAAI Press, 2024. [doi]

Abstract

Abstract is missing.