PinnedAntoine LabatieinTowards Data ScienceIt’s Necessary to Combine Batch Norm and Skip ConnectionsThese techniques must go hand in handAug 31, 2019Aug 31, 2019
Antoine LabatieinTowards Data ScienceRemoving Batch Dependence in CNNs by Proxy-Normalizing ActivationsMemory-efficient convolutional neural network trainingJun 24, 2021Jun 24, 2021
Antoine LabatieinTowards Data ScienceThe Serendipitous Effectiveness of Weight Decay in Deep LearningComing full circle after unintended change of effectsDec 30, 2020Dec 30, 2020