Can Adversarial Training Be Manipulated By Non-Robust Features?
Lue TaoLei FengHongxin WeiJinfeng YiSheng-Jun HuangSongcan ChenPublished in: NeurIPS (2022)
Keyphrases
- feature set
- rotational invariant
- training algorithm
- benchmark datasets
- image features
- feature space
- feature extraction
- multi agent
- low level
- robust estimation
- training examples
- spatial information
- neural network
- training and testing data
- salient features
- learning algorithm
- training process
- feature detection
- structural information
- feature selection
- e learning
- classification accuracy
- prior knowledge
- feature vectors