Secure Split Learning against Property Inference, Data Reconstruction, and Feature Space Hijacking Attacks.
Yunlong MaoZexi XinZhenyu LiJue HongQingyou YangSheng ZhongPublished in: CoRR (2023)
Keyphrases
- data points
- data sets
- feature space
- training data
- database
- prior knowledge
- data analysis
- original data
- learning process
- supervised learning
- data quality
- data sources
- active learning
- learning algorithm
- input data
- labeled data
- high dimensional data
- background knowledge
- data mining
- bayesian networks
- dynamic bayesian networks
- security mechanisms
- structured prediction
- inference process