mlsanityData to 14 Oct 2026

Privacy-Aware Compression for Federated Learning Through Numerical Mechanism Design

arXiv 2211.03942 ↗First posted 8 Nov 2022cs.LG cs.CRcited by 6

Not on the map. Why?

Abstract (arXiv)

In private federated learning (FL), a server aggregates differentially private updates from a large number of clients in order to train a machine learning model. The main challenge in this setting is balancing privacy with both classification accuracy of the learnt model as well as the number of bits communicated between the clients and server. Prior work has achieved a good trade-off by designing a privacy-aware compression mechanism, called the minimum variance unbiased (MVU) mechanism, that numerically solves an optimization problem to determine the parameters of the mechanism. This paper builds upon it by introducing a new interpolation procedure in the numerical design process that allows for a far more efficient privacy analysis. The result is the new Interpolated MVU mechanism that is more scalable, has a better privacy-utility trade-off, and provides SOTA results on communication-efficient private FL on a variety of datasets.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 40 of its 53 references that are in the map. First posted on arXiv 8 Nov 2022.

Builds on

40 of 53 references.

References · 40

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  2. 2016Wide Residual Networkscited by 3,070
  3. 2019Advances and Open Problems in Federated Learningcited by 3,046
  4. 2016Deep Learning with Differential Privacycited by 2,728
  5. 2019Deep Leakage from Gradientscited by 1,021
  6. 2018LEAF: A Benchmark for Federated Settingscited by 792
  7. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 712
  8. 2017Renyi Differential Privacycited by 650
  9. 2017Differentially Private Federated Learning: A Client Level Perspectivecited by 459
  10. 2020iDLG: Improved Deep Leakage from Gradientscited by 382
  11. 2019FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantizationcited by 356
  12. 2021See through Gradients: Image Batch Recovery via GradInversioncited by 326
28 more
  1. 2020Differentially Private Learning Needs Better Features (or Much More Data)cited by 235
  2. 2019Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computationscited by 181
  3. 2020Federated Learning with Compression: Unified Analysis and Sharp Guaranteescited by 160
  4. 2022Reconstructing Training Data with Informed Adversariescited by 148
  5. 2020The Discrete Gaussian for Differential Privacycited by 135
  6. 2018Protection Against Reconstruction and Its Applications in Private Federated Learningcited by 130
  7. 2021The Distributed Discrete Gaussian Mechanism for Federated Learning with Secure Aggregationcited by 128
  8. 2021Linear Convergence in Federated Learning: Tackling Client Heterogeneity and Sparse Gradientscited by 123
  9. 2021Gradient Inversion with Generative Image Priorcited by 101
  10. 2019Hypothesis Testing Interpretations and Renyi Differential Privacycited by 85
  11. 2020Breaking the Communication-Privacy-Accuracy Trilemmacited by 68
  12. 2021The Skellam Mechanism for Differentially Private Federated Learningcited by 68
  13. 2020Bidirectional compression in heterogeneous settings for distributed or federated learning with partial participation: tight convergence guaranteescited by 46
  14. 2022Bounding Training Data Reconstruction in Private (Deep) Learningcited by 46
  15. 2020Optimal Gradient Compression for Distributed and Federated Learningcited by 39
  16. 2022The Fundamental Price of Secure Aggregation in Differentially Private Federated Learningcited by 39
  17. 2022Optimal Algorithms for Mean Estimation under Local Differential Privacycited by 36
  18. 2020Stochastic-Sign SGD for Federated Learning with Theoretical Guaranteescited by 35
  19. 2022Joint Privacy Enhancement and Quantization in Federated Learningcited by 35
  20. 2021Measuring Data Leakage in Machine-Learning Models with Fisher Informationcited by 33
  21. 2022Defending against Reconstruction Attacks with R'enyi Differential Privacycited by 32
  22. 2021Lossless Compression of Efficient Private Local Randomizerscited by 27
  23. 2021Optimal Compression of Locally Differentially Private Mechanismscited by 22
  24. 2022Federated Optimization Algorithms with Random Reshuffling and Gradient Compressioncited by 18
  25. 2022Privacy-Aware Compression for Federated Data Analysiscited by 17
  26. 2022CompressedScaffnew: The First Theoretical Double Acceleration of Communication from Local Training and Compression in Distributed Optimizationcited by 11
  27. 2021DP-REC: Private & Communication-Efficient Federated Learningcited by 10
  28. 2014Optimal Geo-Indistinguishable Mechanisms for Location Privacycited by 9

Built on by

Cited by year

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

6 papers.

  1. 2024Universal Exact Compression of Differentially Private Mechanismscited by 8
  2. 2024Exactly Minimax-Optimal Locally Differentially Private Samplingcited by 5
  3. 2024TernaryVote: Differentially Private, Communication Efficient, and Byzantine Resilient Distributed Optimization on Heterogeneous Datacited by 2
  4. 2024Privacy-Aware Randomized Quantization via Linear Programmingcited by 1
  5. 2024On Using Secure Aggregation in Differentially Private Federated Learning with Multiple Local Stepscited by 1
  6. 2025PBM-VFL: Vertical Federated Learning with Feature and Sample Privacycited by 0