arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2305.18355cs.SDcs.AIcs.LGeess.AS
收藏

基于近端初始化的扩散模型高效成员推理攻击

An Efficient Membership Inference Attack for the Diffusion Model by Proximal Initialization

University of Electronic Science and Technology of China(电子科技大学);Drexel University(德雷塞尔大学)

Fei Kong, Jinhao Duan, RuiPeng Ma, Hengtao Shen, Xiaofeng Zhu, Xiaoshuang Shi, Kaidi Xu

AI总结:

该研究提出基于近端初始化的高效成员推理攻击方法PIA,仅需2次查询即可在扩散模型上取得有竞争力的攻击效果,还首次探索了文本转语音任务中扩散模型的成员推理攻击鲁棒性。

摘要:

近年来,扩散模型在图像、音频生成等生成任务中取得了显著成效,但和其他生成模型一样,扩散模型也存在隐私风险。本文提出一种高效的基于查询的成员推理攻击(MIA),名为近端初始化攻击(PIA),该方法利用t=0时刻由ε初始化得到的真实轨迹与预测点来推断成员身份。实验结果表明,该方法仅需2次查询,就能在离散时间和连续时间扩散模型上取得具有竞争力的性能。此外,此前扩散模型隐私相关的研究多聚焦于视觉任务,未涉及音频任务,因此本文还探索了文本转语音(TTS)这一音频生成任务中扩散模型对MIA的鲁棒性,据我们所知,这是首个研究TTS任务中扩散模型MIA鲁棒性的工作。实验结果显示,输出梅尔频谱图(类图像形式)的模型易受MIA攻击,而输出音频的模型对MIA相对鲁棒。代码可在https://github.com/kong13661/PIA获取。

英文摘要:

Recently, diffusion models have achieved remarkable success in generating tasks, including image and audio generation. However, like other generative models, diffusion models are prone to privacy issues. In this paper, we propose an efficient query-based membership inference attack (MIA), namely Proximal Initialization Attack (PIA), which utilizes groundtruth trajectory obtained by $ε$ initialized in $t=0$ and predicted point to infer memberships. Experimental results indicate that the proposed method can achieve competitive performance with only two queries on both discrete-time and continuous-time diffusion models. Moreover, previous works on the privacy of diffusion models have focused on vision tasks without considering audio tasks. Therefore, we also explore the robustness of diffusion models to MIA in the text-to-speech (TTS) task, which is an audio generation task. To the best of our knowledge, this work is the first to study the robustness of diffusion models to MIA in the TTS task. Experimental results indicate that models with mel-spectrogram (image-like) output are vulnerable to MIA, while models with audio output are relatively robust to MIA. {Code is available at \url{https://github.com/kong13661/PIA}}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑