11-30

(2019) Randaugment:Practical data augmentation with no separate search

11-30

(ICLR 2019) Imagenet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness

11-30

(2019) Fixing the train-test resolution discrepancy

11-23

(ECCV 2016) Deep networks with stochastic depth

01-12

(NIPS 2018 Spotlight) Training Neural Networks Using Features Replay