P-SGD: A Stochastic Gradient Descent Solution for Privacy-Preserving During Protection Transitions.
Karam Bou ChaayaRichard ChbeirMahmoud BarhamgiPhilippe ArnouldDjamal BenslimanePublished in: CAiSE (2021)
Keyphrases
- privacy preserving
- stochastic gradient descent
- privacy guarantees
- privacy protection
- loss function
- privacy preserving data mining
- least squares
- step size
- matrix factorization
- privacy preservation
- random forests
- differential privacy
- vertically partitioned data
- data privacy
- sensitive information
- record linkage
- secure multiparty computation
- online algorithms
- private information
- regularization parameter
- support vector machine
- privacy concerns
- sensitive data
- multiple kernel learning
- weight vector
- importance sampling
- pairwise
- data quality
- logistic regression
- model selection
- linear svm
- feature vectors
- support vector
- objective function