Adversarial Examples Improve Image Recognition

Cihang XieMingxing TanBoqing GongJiang WangAlan YuilleQuoc V. Le

   Papers with code   Abstract  PDF

Adversarial examples are commonly viewed as a threat to ConvNets. Here we present an opposite perspective: adversarial examples can be used to improve image recognition models if harnessed in the right manner... (read more)

Benchmarked Models

RANK
MODEL
REPO
CODE RESULT
PAPER RESULT
ε-REPRODUCED
BUILD
1
EfficientNet-B8
(AdvProp)
85.4%
--
2
EfficientNet-B7
(AdvProp)
85.1%
--
3
EfficientNet-B6
(AdvProp)
84.8%
--
4
EfficientNet-B5
(AdvProp)
84.3%
--
5
EfficientNet-B4
(AdvProp)
83.2%
--
6
EfficientNet-B3
(AdvProp)
81.8%
--
7
EfficientNet-B2
(AdvProp)
80.3%
--
8
EfficientNet-B1
(AdvProp)
79.3%
--
9
EfficientNet-B0
(AdvProp)
77.1%
--