Boosting Gradient for White-Box Adversarial Attacks.
Hongying LiuZhenyu ZhouFanhua ShangXiaoyu QiYuanyuan LiuLicheng JiaoPublished in: CoRR (2020)
Keyphrases
- white box
- black box
- source code
- weighted sums
- countermeasures
- test cases
- learning algorithm
- multi agent
- source code metrics
- ensemble methods
- ensemble learning
- malicious attacks
- traffic analysis
- test data
- security threats
- edge detection
- training data
- feature selection
- watermarking scheme
- machine learning
- watermarking algorithm
- watermarking technique
- weak learners
- malicious users
- terrorist attacks
- chosen plaintext
- base classifiers
- dos attacks
- gradient information
- security mechanisms
- data sets
- cost sensitive
- test set
- relational databases
- training set
- support vector
- real world