Proper Network Interpretability Helps Adversarial Robustness in Classification.
Akhilan BoopathySijia LiuGaoyuan ZhangCynthia LiuPin-Yu ChenShiyu ChangLuca DanielPublished in: ICML (2020)
Keyphrases
- pattern recognition
- pattern classification
- classification accuracy
- classification scheme
- machine learning
- support vector machine
- automatic classification
- support vector
- preprocessing
- feature vectors
- radial basis function network
- decision trees
- image classification
- text classification
- machine learning algorithms
- benchmark datasets
- classification systems
- network model
- network topologies
- network structure
- classification rules
- machine learning methods
- classification method
- classification algorithm
- prediction accuracy
- unsupervised learning
- feature extraction
- network traffic
- rule sets
- training samples
- artificial neural networks
- training set
- feature space
- feature selection