Perturbation of convex risk minimization and its application in differential private learning algorithms

DOI10.1186/S13660-016-1280-0zbMATH Open1359.90096OpenAlexW2569771790WikidataQ37562838 ScholiaQ37562838MaRDI QIDQ504548FDOQ504548

Authors: Weilin Nie, Cheng Wang 

Publication date: 17 January 2017

Published in: Journal of Inequalities and Applications (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1186/s13660-016-1280-0

Recommendations

• Differentially private empirical risk minimization
• Differentially private distributed logistic regression with the objective function perturbation
• Differentially private SGD with non-smooth losses
• Privacy aware learning
• A novel adaptive differential privacy algorithm for empirical risk minimization

zbMATH Keywords

concentration inequalityperturbationdifferential privacyerror decompositionconvex risk minimization

Mathematics Subject Classification ID

Convex programming (90C25)

Cites Work

• Differential Privacy
• Learning Theory
• Probability Inequalities for Sums of Bounded Random Variables
• 10.1162/153244302760200704
• On the mathematical foundations of learning
• Title not available (Why is that?)
• Support vector machine soft margin classifiers: error analysis
• Differential Privacy: A Survey of Results
• Differentially private empirical risk minimization
• Theory of Cryptography
• Learning theory estimates for coefficient-based regularized regression
• Conditional quantiles with varying Gaussians
• Learning with sample dependent hypothesis spaces
• Learning theory estimates via integral operators and their approximations
• Approximation analysis of learning algorithms for support vector regression and quantile regression
• Optimal learning rates for least squares regularized regression with unbounded sampling
• Estimating conditional quantiles with the help of the pinball loss
• Learnability, stability and uniform convergence
• Learning with differential privacy: stability, learnability and the sufficiency and necessity of ERM principle
• Preserving statistical validity in adaptive data analysis (extended abstract)
• Algorithmic stability for adaptive data analysis

Cited In (3)

• \(\mathcal {JHR}\)-operator pairs in \(C^{*}\)-algebra-valued modular metric spaces and related fixed point results via \(C_{*}\)-class functions
• Differentially private distributed logistic regression with the objective function perturbation
• Differentially private SGD with non-smooth losses

Uses Software

• PrivateLR

This page was built for publication: Perturbation of convex risk minimization and its application in differential private learning algorithms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q504548)