Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation

Abstract

Diffusion models have demonstrated remarkable performance in image generation tasks while also raising security and privacy concerns. To tackle these issues, we propose a method for generating unlearnable examples for diffusion models, Unlearnable Diffusion Perturbation, to safeguard images from unauthorized exploitation. Our approach involves designing an algorithm to generate sample-wise perturbation noise for each image to be protected. We frame this as a max-min optimization problem and introduce EUDP, a noise scheduler-based method to enhance the effectiveness of the protective noise. Our experiments demonstrate that training diffusion models on the protected data leads to a significant reduction in the quality of the generated images.

Cite

Text

Zhao et al. "Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation." ICLR 2024 Workshops: R2-FM, 2024.

Markdown

[Zhao et al. "Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation." ICLR 2024 Workshops: R2-FM, 2024.](https://mlanthology.org/iclrw/2024/zhao2024iclrw-unlearnable/)

BibTeX

@inproceedings{zhao2024iclrw-unlearnable,
  title     = {{Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation}},
  author    = {Zhao, Zhengyue and Duan, Jinhao and Hu, Xing and Xu, Kaidi and Wang, Chenan and Zhang, Rui and Du, Zidong and Guo, Qi and Chen, Yunji},
  booktitle = {ICLR 2024 Workshops: R2-FM},
  year      = {2024},
  url       = {https://mlanthology.org/iclrw/2024/zhao2024iclrw-unlearnable/}
}