mlsanityData to 14 Oct 2026

Samplable Anonymous Aggregation for Private Federated Data Analysis

arXiv 2307.15017 ↗First posted 27 Jul 2023cs.CR cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

We revisit the problem of designing scalable protocols for private statistics and private federated learning when each device holds its private data. Locally differentially private algorithms require little trust but are (provably) limited in their utility. Centrally differentially private algorithms can allow significantly better utility but require a trusted curator. This gap has led to significant interest in the design and implementation of simple cryptographic primitives, that can allow central-like utility guarantees without having to trust a central server. Our first contribution is to propose a new primitive that allows for efficient implementation of several commonly used algorithms, and allows for privacy accounting that is close to that in the central setting without requiring the strong trust assumptions it entails. em Shuffling and em aggregation primitives that have been proposed in earlier works enable this for some algorithms, but have significant limitations as primitives. We propose a em Samplable Anonymous Aggregation primitive, which computes an aggregate over a random subset of the inputs and show that it leads to better privacy-utility trade-offs for various fundamental tasks. Secondly, we propose a system architecture that implements this primitive and perform a security analysis of the proposed system. Our design combines additive secret-sharing with anonymization and authentication infrastructures.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 68 of its 155 references that are in the map. First posted on arXiv 27 Jul 2023.

Builds on

68 of 155 references.

References · 68

  1. 2016Deep Learning with Differential Privacycited by 2,728
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  3. 2020Extracting Training Data from Large Language Modelscited by 2,189
  4. 2019Deep Leakage from Gradientscited by 1,021
  5. 2018The Secret Sharer: Evaluating and Testing Unintended Memorization in Neural Networkscited by 864
  6. 2018How To Backdoor Federated Learningcited by 765
  7. 2022Quantifying Memorization Across Neural Language Modelscited by 763
  8. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 712
  9. 2017Renyi Differential Privacycited by 650
  10. 2018signSGD: Compressed Optimisation for Non-Convex Problemscited by 571
  11. 2018Exploiting Unintended Feature Leakage in Collaborative Learningcited by 520
  12. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 497
38 more
  1. 2019Does Learning Require Memorization? A Short Tale about a Long Tailcited by 460
  2. 2021Opacus: User-Friendly Differential Privacy Library in PyTorchcited by 383
  3. 2020iDLG: Improved Deep Leakage from Gradientscited by 382
  4. 2018Analyzing Federated Learning through an Adversarial Lenscited by 351
  5. 2016Concentrated Differential Privacy: Simplifications, Extensions, and Lower Boundscited by 346
  6. 2021See through Gradients: Image Batch Recovery via GradInversioncited by 326
  7. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 309
  8. 2019A Little Is Enough: Circumventing Defenses For Distributed Learningcited by 288
  9. 2014RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Responsecited by 272
  10. 2019Can You Really Backdoor Federated Learning?cited by 255
  11. 2018Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learningcited by 246
  12. 2018Subsampled R'enyi Differential Privacy and Analytical Moments Accountantcited by 236
  13. 2021Adversarial Neuron Pruning Purifies Backdoored Deep Modelscited by 235
  14. 2018Improving the Gaussian Mechanism for Differential Privacy: Analytical Calibration and Optimal Denoisingcited by 218
  15. 2019R'enyi Differential Privacy of the Sampled Gaussian Mechanismcited by 204
  16. 2021Practical and Private (Deep) Learning without Sampling or Shufflingcited by 170
  17. 2018Amplification by Shuffling: From Local to Central Differential Privacy via Anonymitycited by 164
  18. 2017Collecting Telemetry Data Privatelycited by 149
  19. 2020The Discrete Gaussian for Differential Privacycited by 135
  20. 2021When the Curious Abandon Honesty: Federated Learning Is Not Privatecited by 127
  21. 2021Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Production Federated Learningcited by 118
  22. 2018Distributed Differential Privacy via Shufflingcited by 115
  23. 2021Robbing the Fed: Directly Obtaining Private Data in Federated Learning with Modified Modelscited by 108
  24. 2018Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardwarecited by 102
  25. 2020Hiding Among the Clones: A Simple and Nearly Optimal Analysis of Privacy Amplification by Shufflingcited by 96
  26. 2021Papaya: Practical, Private, and Scalable Federated Learningcited by 90
  27. 2021Federated Evaluation and Tuning for On-Device Personalization: System Design & Applicationscited by 85
  28. 2020Turbo-Aggregate: Breaking the Quadratic Aggregation Barrier in Secure Federated Learningcited by 78
  29. 2019The Privacy Blanket of the Shuffle Modelcited by 74
  30. 2022Fishing for User Data in Large-Batch Federated Learning via Gradient Magnificationcited by 73
  31. 2021Eluding Secure Aggregation in Federated Learning via Model Inconsistencycited by 67
  32. 2023Federated Learning of Gboard Language Models with Differential Privacycited by 66
  33. 2017Prochlo: Strong Privacy for Analytics in the Crowdcited by 51
  34. 2022Geographic Spines in the 2020 Census Disclosure Avoidance Systemcited by 44
  35. 2017Prio: Private, Robust, and Scalable Computation of Aggregate Statisticscited by 36
  36. 2020Private Summation in the Multi-Message Shuffle Modelcited by 34
  37. 2019Federated Heavy Hitters Discovery with Differential Privacycited by 32
  38. 2020Improving on-device speaker verification using federated learning with privacycited by 30

50 most cited of 68 All 68 (JSON)

Built on by

Cited by year

Citing papers per year: 2024: 5; 2025: 1.

6 papers.

  1. 2024pfl-research: simulation framework for accelerating research in Private Federated Learningcited by 8
  2. 2024Confidential Federated Computationscited by 4
  3. 2024Momentum Approximation in Asynchronous Private Federated Learningcited by 3
  4. 2024Distributed Differentially Private Data Analytics via Secure Sketchingcited by 1
  5. 2024PAPAYA Federated Analytics Stack: Engineering Privacy, Scalability and Practicalitycited by 1
  6. 2025PREAMBLE: Private and Efficient Aggregation via Block Sparse Vectorscited by 1