mlsanityData to 7 Oct 2026

Membership Privacy Protection for Image Translation Models via Adversarial Knowledge Distillation

arXiv 2203.05212 ↗First posted 10 Mar 2022cs.CV cs.CRcited by 0

On the map: LLM safety and security → Membership inference attacks

Abstract (arXiv)

Image-to-image translation models are shown to be vulnerable to the Membership Inference Attack (MIA), in which the adversary's goal is to identify whether a sample is used to train the model or not. With daily increasing applications based on image-to-image translation models, it is crucial to protect the privacy of these models against MIAs. We propose adversarial knowledge distillation (AKD) as a defense method against MIAs for image-to-image translation models. The proposed method protects the privacy of the training samples by improving the generalizability of the model. We conduct experiments on the image-to-image translation models and show that AKD achieves the state-of-the-art utility-privacy tradeoff by reducing the attack performance up to 38.9% compared with the regular training model at the cost of a slight drop in the quality of the generated output images. The experimental results also indicate that the models trained by AKD generalize better than the regular training models. Furthermore, compared with existing defense methods, the results show that at the same privacy protection level, image translation models trained by AKD generate outputs with higher quality; while at the same quality of outputs, AKD enhances the privacy protection over 30%.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Membership inference attacks, one of 12 directions in LLM safety and security. Builds on 22 of its 35 references that are in the map. First posted on arXiv 10 Mar 2022.

Builds on

22 of 35 references.

Closest ancestors · 2

Same direction, newest first

  1. 2021Mitigating Membership Inference Attacks by Self-Distillation Through a Novel Ensemble Architecture15 Oct
  2. 2021Membership Inference Attacks are Easier on Difficult Problems15 Feb

Other references · 20

  1. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  2. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  3. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  4. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  5. 2016Deep Learning with Differential Privacycited by 2,682
  6. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  7. 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,295
  8. 2018Demystifying MMD GANscited by 1,268
  9. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  10. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
  11. 2018ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Modelscited by 475
  12. 2020Differentiable Augmentation for Data-Efficient GAN Trainingcited by 361
8 more
  1. 2019White-box vs Black-box: Bayes Optimal Strategies for Membership Inferencecited by 265
  2. 2019Reconstruction and Membership Inference Attacks against Generative Modelscited by 122
  3. 2019Adversarial Examples Are a Natural Consequence of Test Error in Noisecited by 114
  4. 2017Knock Knock, Who's There? Membership Inference on Aggregate Location Datacited by 63
  5. 2018Algorithms that Remember: Model Inversion Attacks and Data Protection Lawcited by 52
  6. 2019Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentationcited by 29
  7. 2017Discriminative Region Proposal Adversarial Networks for High-Quality Image-to-Image Translationcited by 8
  8. 2021Towards disease-aware image editing of chest X-rayscited by 2

Built on by

Not cited yet.