mlsanityData to 7 Oct 2026

Variational Model Inversion Attacks

arXiv 2201.10787 ↗First posted 26 Jan 2022cs.LG cs.CRcited by 75

Not on the map. Why?

Abstract (arXiv)

Given the ubiquity of deep neural networks, it is important that these models do not reveal information about sensitive data that they have been trained on. In model inversion attacks, a malicious user attempts to recover the private dataset used to train a supervised neural network. A successful model inversion attack should generate realistic and diverse samples that accurately describe each of the classes in the private dataset. In this work, we provide a probabilistic interpretation of model inversion attacks, and formulate a variational objective that accounts for both diversity and accuracy. In order to optimize this variational objective, we choose a variational family defined in the code space of a deep generative model, trained on a public auxiliary dataset that shares some structural similarity with the target dataset. Empirically, our method substantially improves performance in terms of target attack accuracy, sample realism, and diversity on datasets of faces and chest X-ray images.

Key facts. Cited by 75 papers in the map (data through 7 Oct 2026). Builds on 40 of its 67 references that are in the map. First posted on arXiv 26 Jan 2022.

Builds on

40 of 67 references.

References · 40

  1. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  2. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2017On Calibration of Modern Neural Networkscited by 4,860
  5. 2014Going Deeper with Convolutionscited by 4,231
  6. 2013Auto-Encoding Variational Bayescited by 3,980
  7. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  8. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  9. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  10. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  11. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  12. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,048
28 more
  1. 2016Density estimation using Real NVPcited by 2,029
  2. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  3. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  4. 2016Variational Inference: A Review for Statisticianscited by 1,466
  5. 2014NICE: Non-linear Independent Components Estimationcited by 1,052
  6. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  7. 2019Towards Federated Learning at Scale: System Designcited by 826
  8. 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
  9. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 707
  10. 2019A Closer Look at Few-shot Classificationcited by 679
  11. 2019Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?cited by 675
  12. 2017Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discoverycited by 566
  13. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 425
  14. 2019Dreaming to Distill: Data-free Knowledge Transfer via DeepInversioncited by 419
  15. 2019Federated Machine Learning: Concept and Applicationscited by 409
  16. 2014Understanding Deep Image Representations by Inverting Themcited by 399
  17. 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 393
  18. 2017Deep Models Under the GAN: Information Leakage from Collaborative Deep Learningcited by 326
  19. 2019The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networkscited by 241
  20. 2020Few-Shot Learning via Learning the Representation, Provablycited by 213
  21. 2020On the Theory of Transfer Learning: The Importance of Task Diversitycited by 186
  22. 2013A General Framework for Updating Belief Distributionscited by 172
  23. 2016Semantic Image Inpainting with Deep Generative Modelscited by 156
  24. 2019MineGAN: effective knowledge transfer from GANs to target domains with few imagescited by 106
  25. 2019Generalized Variational Inference: Three arguments for deriving new Posteriorscited by 57
  26. 2017Assigning a value to a power likelihood in a general Bayesian modelcited by 46
  27. 2015Robust Bayesian inference via coarseningcited by 32
  28. 2019Adversarial Neural Network Inversion via Auxiliary Knowledge Alignmentcited by 30

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 22; 2024: 26; 2025: 15; 2026: 6.

75 papers.

Sort:

By area:

  1. 2024Building Guardrails for Large Language Modelscited by 61
  2. 2022Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attackscited by 58
  3. 2023CodeLMSec Benchmark: Systematically Evaluating and Finding Security Vulnerabilities in Black-Box Code Language Modelscited by 52
  4. 2023Re-thinking Model Inversion Attacks Against Deep Neural Networkscited by 39
  5. 2023Reinforcement Learning-Based Black-Box Model Inversion Attackscited by 33
  6. 2022Bilateral Dependency Optimization: Defending Against Model-inversion Attackscited by 30
  7. 2024Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defensescited by 30
  8. 2022Text Revealer: Private Text Reconstruction via Model Inversion Attacks against Transformerscited by 28
  9. 2023Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial Networkcited by 28
  10. 2023Label-Only Model Inversion Attacks via Knowledge Transfercited by 21
  11. 2023FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUscited by 20
  12. 2022Does CLIP Know My Face?cited by 19
38 more
  1. 2024Is Retain Set All You Need in Machine Unlearning? Restoring Performance of Unlearned Models with Out-Of-Distribution Imagescited by 18
  2. 2023Private, fair and accurate: Training large-scale, privacy-preserving AI models in medical imagingcited by 17
  3. 2024A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attackscited by 17
  4. 2023On Strengthening and Defending Graph Reconstruction Attack with Markov Chain Approximationcited by 16
  5. 2023Be Careful What You Smooth For: Label Smoothing Can Be a Privacy Shield but Also a Catalyst for Model Inversion Attackscited by 12
  6. 2024Model Inversion Robustness: Can Transfer Learning Help?cited by 11
  7. 2024Model Inversion Attacks: A Survey of Approaches and Countermeasurescited by 11
  8. 2022Fast Deep Autoencoder for Federated learningcited by 10
  9. 2023BrainWash: A Poisoning Attack to Forget in Continual Learningcited by 9
  10. 2024Improving Robustness to Model Inversion Attacks via Sparse Coding Architecturescited by 8
  11. 2024Model Inversion Attacks Through Target-Specific Conditional Diffusion Modelscited by 8
  12. 2024On the Vulnerability of Skip Connections to Model Inversion Attackscited by 8
  13. 2023Boosting Model Inversion Attacks with Adversarial Examplescited by 7
  14. 2024BrainLeaks: On the Privacy-Preserving Properties of Neuromorphic Architectures against Model Inversion Attackscited by 7
  15. 2025Differential privacy for medical deep learning: methods, tradeoffs, and deployment implicationscited by 7
  16. 2023Model Inversion Attack via Dynamic Memory Learningcited by 6
  17. 2025Theoretical Insights in Model Inversion Robustness and Conditional Entropy Maximization for Collaborative Inference Systemscited by 6
  18. 2024Understanding Data Importance in Machine Learning Attacks: Does Valuable Data Pose Greater Harm?cited by 5
  19. 2024MIBench: A Comprehensive Framework for Benchmarking Model Inversion Attack and Defensecited by 5
  20. 2025SoK: Data Reconstruction Attacks Against Machine Learning Models: Definition, Metrics, and Benchmarkcited by 5
  21. 2023Balancing Transparency and Risk: The Security and Privacy Risks of Open-Source Machine Learning Modelscited by 4
  22. 2023Defending Our Privacy With Backdoorscited by 4
  23. 2024Analyzing Inference Privacy Risks Through Gradients in Machine Learningcited by 4
  24. 2024Trap-MID: Trapdoor-based Defense against Model Inversion Attackscited by 4
  25. 2024Training Data Reconstruction: Privacy due to Uncertainty?cited by 4
  26. 2024Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Datacited by 3
  27. 2024Recent Advances, Applications, and Open Challenges in Machine Learning for Health: Reflections from Research Roundtables at ML4H 2023 Symposiumcited by 3
  28. 2025Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learningcited by 3
  29. 2023Class Attribute Inference Attacks: Inferring Sensitive Class Information by Diffusion-Based Attribute Manipulationscited by 2
  30. 2023Robust Representation Learning for Privacy-Preserving Machine Learning: A Multi-Objective Autoencoder Approachcited by 2
  31. 2023CovarNav: Machine Unlearning via Model Inversion and Covariance Navigationcited by 2
  32. 2024Inference Attacks: A Taxonomy, Survey, and Promising Directionscited by 2
  33. 2024Prediction Exposes Your Face: Black-box Model Inversion via Prediction Alignmentcited by 2
  34. 2023When Machine Learning Models Leak: An Exploration of Synthetic Training Datacited by 1
  35. 2023Reconciling AI Performance and Data Reconstruction Resilience for Medical Imagingcited by 1
  36. 2024Privacy Re-identification Attacks on Tabular GANscited by 1
  37. 2024Knowledge Distillation-Based Model Extraction Attack using GAN-based Private Counterfactual Explanationscited by 1
  38. 2024Distributional Black-Box Model Inversion Attack with Multi-Agent Reinforcement Learningcited by 1

50 most cited of 75 All 75 (JSON)