Labels Are Culprits: Defending Gradient Attack on Privacy.
Zhaohua LiLe WangZhaoquan GuYang LvZhihong TianPublished in: IEEE Internet Things J. (2024)
Keyphrases
- ddos attacks
- security risks
- authentication protocol
- privacy preserving
- training data
- gradient method
- privacy aware
- pairwise
- data sets
- private information
- privacy preserving data mining
- privacy concerns
- privacy preservation
- gradient information
- security issues
- personal data
- statistical databases
- differential privacy
- sensitive information
- privacy protection
- privacy issues
- class labels
- gradient direction
- privacy enhancing
- malicious users
- security measures
- multiscale
- identity management
- training set
- rfid systems
- security properties
- private data
- dos attacks
- multi label
- network security