The training of diffusion-based models for image generation is predominantly controlled by a select few Big Tech companies, raising concerns about privacy, copyright, and data authority due to their lack of transparency regarding training data. To ad-dress this issue, we propose a federated diffusion model scheme that enables the independent and collaborative training of diffusion models without exposing local data. Our approach adapts the Federated Averaging (FedAvg) algorithm to train a Denoising Diffusion Model (DDPM). Through a novel utilization of the underlying UNet backbone, we achieve a significant reduction of up to 74% in the number of parameters exchanged during training,compared to the naive FedAvg approach, whilst simultaneously maintaining image quality comparable to the centralized setting, as evaluated by the FID score.

翻译：基于扩散的图像生成模型训练主要由少数大型科技公司控制，由于其训练数据缺乏透明度，引发了关于隐私、版权和数据所有权的担忧。为解决这一问题，我们提出了一种联邦扩散模型方案，使得扩散模型能够在无需暴露本地数据的情况下进行独立且协作的训练。我们的方法通过调整联邦平均（FedAvg）算法来训练去噪扩散模型（DDPM）。通过对底层UNet主干网络的新颖利用，相较于朴素的FedAvg方法，我们在训练过程中交换的参数数量显著减少了高达74%，同时根据FID分数评估，保持了与集中式训练相当的图像质量。