MaxUp: A Simple Way to Improve Generalization of Neural Network Training

20 Feb 2020 Chengyue Gong Tongzheng Ren Mao Ye Qiang Liu

We propose \emph{MaxUp}, an embarrassingly simple, highly effective technique for improving the generalization performance of machine learning models, especially deep neural networks. The idea is to generate a set of augmented data with some random perturbations or transforms and minimize the maximum, or worst case loss over the augmented data... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Classification ImageNet Fix-EfficientNet-B8 (MaxUp + CutMix) Top 1 Accuracy 85.8% # 12
Number of params 87.42M # 15

Methods used in the Paper