mlsanityData to 7 Oct 2026

InstaHide: Instance-hiding Schemes for Private Distributed Learning

arXiv 2010.02772 ↗First posted 6 Oct 2020cs.CR cs.CC cs.DScited by 90

Not on the map. Why?

Abstract (arXiv)

How can multiple distributed entities collaboratively train a shared deep net on their private data while preserving privacy? This paper introduces InstaHide, a simple encryption of training images, which can be plugged into existing distributed deep learning pipelines. The encryption is efficient and applying it during training has minor effect on test accuracy. InstaHide encrypts each training image with a "one-time secret key" which consists of mixing a number of randomly chosen images and applying a random pixel-wise mask. Other contributions of this paper include: (a) Using a large public dataset (e.g. ImageNet) for mixing during its encryption, which improves security. (b) Experimental results to show effectiveness in preserving privacy against known attacks with only minor effects on accuracy. (c) Theoretical analysis showing that successfully attacking privacy requires attackers to solve a difficult computational problem. (d) Demonstrating that use of the pixel-wise mask is important for security, since Mixup alone is shown to be insecure to some some efficient attacks. (e) Release of a challenge dataset https://github.com/Hazelsuko07/InstaHide_Challenge Our code is available at https://github.com/Hazelsuko07/InstaHide

Key facts. Cited by 90 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Oct 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 18; 2022: 27; 2023: 18; 2024: 11; 2025: 7; 2026: 9.

90 papers.

Sort:

By area:

  1. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 405
  2. 2021Evaluating Gradient Inversion Attacks and Defenses in Federated Learningcited by 196
  3. 2022Reconstructing Training Data from Trained Neural Networkscited by 129
  4. 2022Towards Adversarial Evaluations for Inexact Machine Unlearningcited by 82
  5. 2022Recovering Private Text in Federated Learning of Language Modelscited by 67
  6. 2021Quantum federated learning through blind quantum computingcited by 44
  7. 2022A Survey on Gradient Inversion: Attacks, Defenses and Future Directionscited by 39
  8. 2022Gradient Obfuscation Gives a False Sense of Security in Federated Learningcited by 37
  9. 2021CaPC Learning: Confidential and Private Collaborative Learningcited by 33
  10. 2021DP-InstaHide: Provably Defusing Poisoning and Backdoor Attacks with Differentially Private Data Augmentationscited by 33
  11. 2022Privacy-Preserving Face Recognition with Learnable Privacy Budgets in Frequency Domaincited by 28
  12. 2023Privacy and Fairness in Federated Learning: on the Perspective of Trade-offcited by 26
38 more
  1. 2021Survey: Image Mixing and Deleting for Data Augmentationcited by 25
  2. 2024Privacy-Preserving Face Recognition Using Trainable Feature Subtractioncited by 24
  3. 2023TransFace++: Rethinking the Face Recognition Paradigm with a Focus on Accuracy, Efficiency, and Securitycited by 22
  4. 2021Towards Understanding the Data Dependency of Mixup-style Trainingcited by 19
  5. 2022DuetFace: Collaborative Privacy-Preserving Face Recognition via Channel Splitting in the Frequency Domaincited by 18
  6. 2022Privacy-Preserving Image Classification Using Isotropic Networkcited by 17
  7. 2022Distributed Contrastive Learning for Medical Image Segmentationcited by 16
  8. 2024Efficiently Serving Large Multimodal Models Using EPD Disaggregationcited by 16
  9. 2021Decentralized Unsupervised Learning of Visual Representationscited by 15
  10. 2023Privacy-Preserving Face Recognition Using Random Frequency Componentscited by 15
  11. 2022FedLite: A Scalable Approach for Federated Learning on Resource-constrained Clientscited by 14
  12. 2022Provably Learning Diverse Features in Multi-View Data with Midpoint Mixupcited by 14
  13. 2022DarKnight: An Accelerated Framework for Privacy and Integrity Preserving Deep Learning Using Trusted Hardwarecited by 13
  14. 2023Optimizing Privacy, Utility and Efficiency in Constrained Multi-Objective Federated Learningcited by 12
  15. 2021Opportunistic Federated Learning: An Exploration of Egocentric Collaboration for Pervasive Computing Applicationscited by 11
  16. 2021NeuraCrypt: Hiding Private Health Data via Random Neural Networks for Public Trainingcited by 11
  17. 2022Dropout is NOT All You Need to Prevent Gradient Leakagecited by 11
  18. 2022Concealing Sensitive Samples against Gradient Leakage in Federated Learningcited by 11
  19. 2023Bounding the Invertibility of Privacy-preserving Instance Encoding using Fisher Informationcited by 11
  20. 2023Unmasking Transformers: A Theoretical Approach to Data Recovery via Attention Weightscited by 11
  21. 2023FedPass: Privacy-Preserving Vertical Federated Deep Learning with Adaptive Obfuscationcited by 9
  22. 2023Federated Empirical Risk Minimization via Second-Order Methodcited by 9
  23. 2021A Method to Reveal Speaker Identity in Distributed ASR Training, and How to Counter Itcited by 8
  24. 2023Expressive variational quantum circuits provide inherent privacy in federated learningcited by 8
  25. 2021Disrupting Model Training with Adversarial Shortcutscited by 7
  26. 2021Practical and Secure Federated Recommendation with Personalized Maskscited by 7
  27. 2022Deploying Convolutional Networks on Untrusted Platforms Using 2D Holographic Reduced Representationscited by 7
  28. 2024Fast John Ellipsoid Computation with Differential Privacy Optimizationcited by 7
  29. 2022FedDef: Defense Against Gradient Leakage in Federated Learning-based Network Intrusion Detection Systemscited by 6
  30. 2023Understanding Deep Gradient Leakage via Inversion Influence Functionscited by 6
  31. 2025Uncovering Gradient Inversion Risks in Practical Language Model Trainingcited by 6
  32. 2022Enhancing Privacy against Inversion Attacks in Federated Learning by using Mixing Gradients Strategiescited by 5
  33. 2021A Fusion-Denoising Attack on InstaHide with Data Augmentationcited by 4
  34. 2022Differentially Private Deep Learning with ModelMixcited by 4
  35. 2022GAN-based Domain Inference Attackcited by 4
  36. 2023Privacy-Preserving Remote Heart Rate Estimation from Facial Videoscited by 4
  37. 2023Temporal Gradient Inversion Attacks with Robust Optimizationcited by 4
  38. 2023PPIDSG: A Privacy-Preserving Image Distribution Sharing Scheme with GAN in Federated Learningcited by 4

50 most cited of 90 All 90 (JSON)