DIA: Diffusion based Inverse Network Attack on Collaborative Inference

Dake Chen, Shiduo Li, Yuke Zhang, Souvik Kundu, Chenghao Li, Peter Anthony Beerel

Keywords: Diffusion model, data privacy, inverse network attack, collaborative Inference

TL;DR: This paper introduces a diffusion-based inverse network attack, named DIA, for collaborative inference systems.

Abstract: With the continuous expansion of neural networks in size and depth, and the growing popularity of machine learning as a service, collaborative inference systems present a promising approach for deploying models in resource-constrained computing environments. However, as the deployment of these systems gains traction, evaluating their privacy and security has become a critical issue. Towards this goal, this paper introduces a diffusion-based inverse network attack, named DIA, for collaborative inference systems that uses a novel feature map awareness conditioning mechanism to guide the diffusion model training. Compared to prior approaches, our extensive empirical results demonstrate that the proposed attack achieves an average improvement of 29%, 20%, 30% in terms of SSIM, PSNR, and MSE when applied to convolutional neural networks (CNN), 18%, 17%, 61% to ResNet models, and 55%, 54%, 84% to Vision transformers (ViTs). Moreover, we uncover a notable vulnerability of transformer-based model ViTs and analyze the potential reasons behind this vulnerability. Based on our analysis, we raise caution regarding the deployment of transformer-based models in collaborative inference systems, emphasizing the need for careful consideration regarding the security of such models in collaborative settings.

Primary Area: societal considerations including fairness, safety, privacy

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 4538