Differentially private stochastic gradient descent via compression and memorization.
Le Trieu PhongTran Thi PhuongPublished in: J. Syst. Archit. (2023)
Keyphrases
- differentially private
- stochastic gradient descent
- least squares
- differential privacy
- loss function
- matrix factorization
- step size
- support vector machine
- random forests
- regularization parameter
- multiple kernel learning
- weight vector
- privacy preserving
- image compression
- recommender systems
- importance sampling
- online algorithms
- cost function