Concept-based Adversarial Attacks: Tricking Humans and Classifiers Alike.
Johannes SchneiderGiovanni ApruzzesePublished in: CoRR (2022)
Keyphrases
- training data
- support vector
- countermeasures
- classification systems
- decision trees
- text categorization
- supervised classification
- classification models
- classification algorithm
- multiple classifiers
- text clustering
- linear classifiers
- machine learning methods
- svm classifier
- machine learning algorithms
- training examples
- naive bayes
- training set
- bayesian networks
- test set
- human behavior
- human users
- digital images
- multi agent
- majority voting
- malicious attacks