Prochlo: Strong Privacy for Analytics in the Crowd

Abstract

The large-scale monitoring of computer users' software activities has become commonplace, e.g., for application telemetry, error reporting, or demographic profiling. This paper describes a principled systems architecture---Encode, Shuffle, Analyze (ESA)---for performing such monitoring with high utility while also protecting user privacy. The ESA design, and its Prochlo implementation, are informed by our practical experiences with an existing, large deployment of privacy-preserving software monitoring. (cont.; see the paper)

References in corpus (2)

• Deep Learning with Differential Privacy
• RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response

Cited by in corpus (24)

• Distributed Differential Privacy via Shuffling
• How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacy
• Conclave: secure multi-party computation on big data (extended TR)
• Private Summation in the Multi-Message Shuffle Model
• Challenges of AI in Wireless Networks for IoT
• The Internet of Federated Things (IoFT): A Vision for the Future and In-depth Survey of Data-driven Approaches for Federated Learning
• Computing Local Sensitivities of Counting Queries with Joins
• CheckDP: An Automated and Integrated Approach for Proving Differential Privacy or Finding Precise Counterexamples
• Confidentiality Protection in the 2020 US Census of Population and Housing
• STAR: Secret Sharing for Private Threshold Aggregation Reporting
• Dordis: Efficient Federated Learning with Dropout-Resilient Differential Privacy
• Aggregation and Transformation of Vector-Valued Messages in the Shuffle Model of Differential Privacy
• ARA : Aggregated RAPPOR and Analysis for Centralized Differential Privacy
• Network Shuffling: Privacy Amplification via Random Walks
• DPGen: Automated Program Synthesis for Differential Privacy
• Private Prediction Sets
• Local Differentially Private Fuzzy Counting in Stream Data using Probabilistic Data Structure
• Tight Differential Privacy Guarantees for the Shuffle Model with $k$-Randomized Response
• Nebula: Efficient, Private and Accurate Histogram Estimation
• Improving Utility and Security of the Shuffler-based Differential Privacy
• Cookie Monster: Efficient On-device Budgeting for Differentially-Private Ad-Measurement Systems
• Defense against Poisoning Attacks under Shuffle-DP
• On Heuristic Models, Assumptions, and Parameters
• Tight Differential Privacy Blanket for Shuffle Model