mlsanityData to 7 Oct 2026

Representation Magnitude has a Liability to Privacy Vulnerability

arXiv 2407.16164 ↗First posted 23 Jul 2024cs.LG cs.AI cs.CRcited by 3

Not on the map. Why?

Abstract (arXiv)

The privacy-preserving approaches to machine learning (ML) models have made substantial progress in recent years. However, it is still opaque in which circumstances and conditions the model becomes privacy-vulnerable, leading to a challenge for ML models to maintain both performance and privacy. In this paper, we first explore the disparity between member and non-member data in the representation of models under common training frameworks. We identify how the representation magnitude disparity correlates with privacy vulnerability and address how this correlation impacts privacy vulnerability. Based on the observations, we propose Saturn Ring Classifier Module (SRCM), a plug-in model-level solution to mitigate membership privacy leakage. Through a confined yet effective representation space, our approach ameliorates models' privacy vulnerability while maintaining generalizability. The code of this work can be found here: urlhttps://github.com/JEKimLab/AIES2024SRCM

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 37 of its 50 references that are in the map. First posted on arXiv 23 Jul 2024.

Builds on

37 of 50 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  3. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  4. 2017On Calibration of Modern Neural Networkscited by 4,860
  5. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  6. 2019Searching for MobileNetV3cited by 1,984
  7. 2018Group Normalizationcited by 1,514
  8. 2018Obfuscated Gradients Give a False Sense of Security: Circumventing Defenses to Adversarial Examplescited by 1,259
  9. 2016Instance Normalization: The Missing Ingredient for Fast Stylizationcited by 1,187
  10. 2015Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)cited by 1,061
  11. 2019When Does Label Smoothing Help?cited by 985
  12. 2018CosFace: Large Margin Cosine Loss for Deep Face Recognitioncited by 968
25 more
  1. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  2. 2018ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Modelscited by 475
  3. 2017Regularizing Neural Networks by Penalizing Confident Output Distributionscited by 461
  4. 2020Circle Loss: A Unified Perspective of Pair Similarity Optimizationcited by 402
  5. 2022AdaFace: Quality Adaptive Margin for Face Recognitioncited by 390
  6. 2020Label-Only Membership Inference Attackscited by 371
  7. 2021MagFace: A Universal Representation for Face Recognition and Quality Assessmentcited by 332
  8. 2016Large-Margin Softmax Loss for Convolutional Neural Networkscited by 296
  9. 2018Additive Margin Softmax for Face Verificationcited by 275
  10. 2020Systematic Evaluation of Privacy Risks of Machine Learning Modelscited by 273
  11. 2019MemGuard: Defending against Black-Box Membership Inference Attacks via Adversarial Examplescited by 183
  12. 2018Machine Learning with Membership Privacy using Adversarial Regularizationcited by 177
  13. 2020Data-Free Model Extractioncited by 121
  14. 2020MAZE: Data-Free Model Stealing Attack Using Zeroth-Order Gradient Estimationcited by 99
  15. 2020R-GAP: Recursive Gradient Attack on Privacycited by 95
  16. 2021Mitigating Membership Inference Attacks by Self-Distillation Through a Novel Ensemble Architecturecited by 63
  17. 2022Towards Data-Free Model Stealing in a Hard Label Settingcited by 57
  18. 2022Membership Inference Attacks and Defenses in Neural Network Pruningcited by 42
  19. 2022RelaxLoss: Defending Membership Inference Attacks without Losing Utilitycited by 35
  20. 2020Against Membership Inference Attack: Pruning is All You Needcited by 32
  21. 2022Understanding Instance-Level Impact of Fairness Constraintscited by 27
  22. 2021Understanding Training-Data Leakage from Gradients in Neural Networks for Image Classificationcited by 16
  23. 2022Leveraging Adversarial Examples to Quantify Membership Information Leakagecited by 16
  24. 2022Purifier: Defending Data Inference Attacks via Transforming Confidence Scorescited by 15
  25. 2024Center-Based Relaxed Learning Against Membership Inference Attackscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Trustworthy AI: Safety, Bias, and Privacy -- A Surveycited by 5
  2. 2026Decoupling Generalizability and Membership Privacy Risks in Neural Networkscited by 1
  3. 2026Learnability and Privacy Vulnerability are Entangled in a Few Critical Weightscited by 0