A Comprehensive Survey on Local Differential Privacy Toward Data Statistics and Analysis

Abstract

Collecting and analyzing massive data generated from smart devices have become increasingly pervasive in crowdsensing, which are the building blocks for data-driven decision-making. However, extensive statistics and analysis of such data will seriously threaten the privacy of participating users. Local differential privacy (LDP) has been proposed as an excellent and prevalent privacy model with distributed architecture, which can provide strong privacy guarantees for each user while collecting and analyzing data. LDP ensures that each user's data is locally perturbed first in the client-side and then sent to the server-side, thereby protecting data from privacy leaks on both the client-side and server-side. This survey presents a comprehensive and systematic overview of LDP with respect to privacy models, research tasks, enabling mechanisms, and various applications. Specifically, we first provide a theoretical summarization of LDP, including the LDP model, the variants of LDP, and the basic framework of LDP algorithms. Then, we investigate and compare the diverse LDP mechanisms for various data statistics and analysis tasks from the perspectives of frequency estimation, mean estimation, and machine learning. What's more, we also summarize practical LDP-based application scenarios. Finally, we outline several future research directions under LDP.

28pages

References in corpus (63)

• Deep Learning with Differential Privacy
• A Brief Survey of Deep Reinforcement Learning
• RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response
• Geo-Indistinguishability: Differential Privacy for Location-Based Systems
• On the Complexity of Best Arm Identification in Multi-Armed Bandit Models
• Federated Machine Learning: Concept and Applications
• Differentially Private Empirical Risk Minimization
• Differentially Private Federated Learning: A Client Level Perspective
• Local, Private, Efficient Protocols for Succinct Histograms
• LoPub: High-Dimensional Crowdsourced Data Publication with Local Differential Privacy
• Collecting Telemetry Data Privately
• Extremal Mechanisms for Local Differential Privacy
• Protection Against Reconstruction and Its Applications in Private Federated Learning
• Local Differential Privacy for Deep Learning
• Discrete Distribution Estimation under Local Privacy
• Privacy Loss in Apple's Implementation of Differential Privacy on MacOS 10.12
• A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics
• Collecting and Analyzing Data from Smart Device Users with Local Differential Privacy
• Quantifying Differential Privacy under Temporal Correlations
• Hadamard Response: Estimating Distributions Privately, Efficiently, and with Little Communication
• Quantifying Differential Privacy in Continuous Data Release under Temporal Correlations
• Evaluating Differentially Private Machine Learning in Practice
• Mutual Information Optimally Local Private Discrete Distribution Estimation
• Local Differential Privacy and Its Applications: A Comprehensive Survey
• Local Privacy, Data Processing Inequalities, and Statistical Minimax Rates
• Distributed Private Heavy Hitters
• Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing
• Practical Locally Private Heavy Hitters
• Collecting and Analyzing Multidimensional Data with Local Differential Privacy
• Test without Trust: Optimal Locally Private Distribution Testing
• Local Differential Privacy: a tutorial
• LDP-FL: Practical Private Aggregation in Federated Learning with Local Differential Privacy
• Machine Learning Models that Remember Too Much
• Multi-Armed Bandits with Local Differential Privacy
• Clustering Algorithms for the Centralized and Local Models
• LDP-Fed: Federated Learning with Local Differential Privacy
• Wireless Federated Learning with Local Differential Privacy
• Linear Queries Estimation with Local Differential Privacy
• Locally Differentially Private Heavy Hitter Identification
• Local Differential Privacy for Evolving Data
• Differentially Private Meta-Learning
• Corrupt Bandits for Preserving Local Privacy
• Locally Private Distributed Reinforcement Learning
• Adaptive Laplace Mechanism: Differential Privacy Preservation in Deep Learning
• Locally Private Mean Estimation: Z-test and Tight Confidence Intervals
• PCKV: Locally Differentially Private Correlated Key-Value Data Collection with Optimized Utility
• Local Private Hypothesis Testing: Chi-Square Tests
• Utility-Optimized Local Differential Privacy Mechanisms for Distribution Estimation
• Conditional Analysis for Key-Value Data with Local Differential Privacy
• Collect at Once, Use Effectively: Making Non-interactive Locally Private Learning Possible
• Metric-based local differential privacy for statistical applications
• Locally Differentially Private Naive Bayes Classification
• Estimating Smooth GLM in Non-interactive Local Differential Privacy Model with Public Unlabeled Data
• Distributed Clustering in the Anonymized Space with Local Differential Privacy
• Priv'IT: Private and Sample Efficient Identity Testing
• Classification under local differential privacy
• Concentrated Differentially Private Gradient Descent with Adaptive per-Iteration Privacy Budget
• Privacy-preserving Crowd-guided AI Decision-making in Ethical Dilemmas
• Locally Differentially Private Data Collection and Analysis
• FedSel: Federated SGD under Local Differential Privacy with Top-k Dimension Selection
• Locally Private Hypothesis Testing
• Aggregating Votes with Local Differential Privacy: Usefulness, Soundness vs. Indistinguishability
• Parameter-Free Locally Differentially Private Stochastic Subgradient Descent

Cited by in corpus (4)

• Joint Privacy Enhancement and Quantization in Federated Learning
• When differential privacy meets NLP: The devil is in the detail
• Advances in Differential Privacy and Differentially Private Machine Learning
• Bit-efficient Numerical Aggregation and Stronger Privacy for Trust in Federated Analytics