带隐私保证的扩散模型个性化联邦训练
Personalized Federated Training of Diffusion Models with Privacy Guarantees
为在去中心化私有数据上训练扩散模型提供了带逐客户端差分隐私保证的联邦个性化方案，可直接迁移到视觉场景中多源私有图像（如医疗、人脸）的协同生成建模，并给出了条件生成在 GMM 下的效用界作为分析工具。
原文依据 We propose a federated framework for training diffusion models on decentralized and private datasets.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出了一个在去中心化且私有的数据集上训练扩散模型的联邦框架。该方法在训练共享生成模型的同时学习个性化客户端模型，使客户端能够受益于跨客户端结构，同时确保共享模型无法单独复现任何客户端的数据。我们为每个客户端提供了形式化的差分隐私保证，并在高斯混合模型下建立了条件生成的效用界，表明相对于私有的非协作训练，协作能够提升样本质量。在 CIFAR-10、Colorized MNIST 和 CelebA 上的实验支持了这些结果：该方法生成高保真样本，提升了少数类和代表性不足类别的性能，并对成员推断、记忆化和重建攻击保持强防护。
英文原文摘要 863 字符
We propose a federated framework for training diffusion models on decentralized and private datasets. The method learns a shared generative model alongside personalized client models, allowing clients to benefit from cross-client structure while ensuring that the shared model cannot reproduce any client's data on its own. We provide formal differential privacy guarantees for each client and establish utility bounds for conditional generation under a Gaussian mixture model, showing that collaboration improves sample quality relative to private non-collaborative training. Experiments on CIFAR-10, Colorized MNIST, and CelebA support these results: the method generates high-fidelity samples, improves performance on minority and underrepresented classes, and maintains strong protection against membership inference, memorization, and reconstruction attacks.