Privacy Aware Learning

Abstract

We study statistical risk minimization problems under a privacy model in which the data is kept confidential even from the learner. In this local privacy framework, we establish sharp upper and lower bounds on the convergence rates of statistical estimation procedures. As a consequence, we exhibit a precise tradeoff between the amount of privacy the data preserves and the utility, as measured by convergence rate, of any statistical estimator or learning procedure.

60 pages

References in corpus (1)

• Local Privacy, Data Processing Inequalities, and Statistical Minimax Rates

Cited by in corpus (18)

• Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning
• Federated Learning and Differential Privacy: Software tools analysis, the Sherpa.ai FL framework and methodological guidelines for preserving data privacy
• Crypto-Nets: Neural Networks over Encrypted Data
• Federated Deep Reinforcement Learning
• Differential Privacy of Populations in Routing Games
• Differentially Private Testing of Identity and Closeness of Discrete Distributions
• Development and Analysis of Deterministic Privacy-Preserving Policies Using Non-Stochastic Information Theory
• LOGAN: Membership Inference Attacks Against Generative Models
• Privacy and Utility Tradeoff in Approximate Differential Privacy
• Lower complexity bounds of first-order methods for convex-concave bilinear saddle-point problems
• Asynchronous Distributed Learning from Constraints
• Fidelity-Weighted Learning
• Differential Privacy of Aggregated DC Optimal Power Flow Data
• Private Rank Aggregation under Local Differential Privacy
• Learning to Succeed while Teaching to Fail: Privacy in Closed Machine Learning Systems
• Non-Stochastic Hypothesis Testing with Application to Privacy Against Hypothesis-Testing Adversary
• Notes on Information-Theoretic Privacy
• Differentially Private Data Releasing for Smooth Queries with Synthetic Database Output