通过分阶段训练提升私密图像生成质量，先学简单特征再攻复杂细节。
From Easy to Hard: Building a Shortcut for Differentially Private Image Synthesis
- 采用课程学习思路，分两阶段训练：先用聚合图像学简单特征，再生成复杂图像。
- 在4个数据集上，生成图像保真度和实用性分别比当前最优方法高33.1%和2.1%。
- 适合关注隐私保护下图像生成质量的科研人员与工业应用开发者。
差分隐私（DP）图像合成旨在从敏感数据集中生成合成图像，缓解组织共享与使用合成图像时的隐私泄露风险。尽管已有方法在使用差分隐私随机梯度下降（DP-SGD）训练扩散模型方面取得显著进展，但性能仍不理想。受课程学习启发，本文提出一种两阶段的差分隐私图像合成框架，使扩散模型从易到难逐步学习生成私密合成图像。不同于直接使用DP-SGD训练扩散模型的方法，我们引入初始的“简单阶段”，在此阶段利用敏感数据集随机样本的聚合图像（即“中心图像”）进行训练。这些中心图像虽无细节，但能体现整体图像特征，且隐私开销极小，有助于早期模型训练。实验表明，在四个研究的数据集上，本方法生成图像的保真度和实用性平均分别较当前最优方法提升33.1%和2.1%。
原文摘要 · Abstract (English)
Differentially private (DP) image synthesis aims to generate synthetic images from a sensitive dataset, alleviating the privacy leakage concerns of organizations sharing and utilizing synthetic images. Although previous methods have significantly progressed, especially in training diffusion models on sensitive images with DP Stochastic Gradient Descent (DP-SGD), they still suffer from unsatisfactory performance. In this work, inspired by curriculum learning, we propose a two-stage DP image synthesis framework, where diffusion models learn to generate DP synthetic images from easy to hard. Unlike existing methods that directly use DP-SGD to train diffusion models, we propose an easy stage in the beginning, where diffusion models learn simple features of the sensitive images. To facilitate this easy stage, we propose to use `central images', simply aggregations of random samples of the sensitive dataset. Intuitively, although those central images do not show details, they demonstrate useful characteristics of all images and only incur minimal privacy costs, thus helping early-phase model training. We conduct experiments to present that on the average of four investigated image datasets, the fidelity and utility metrics of our synthetic images are 33.1% and 2.1% better than the state-of-the-art method.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。