Federated Class Incremental Learning (FCIL) is a critical yet largely underexplored issue that deals with the dynamic incorporation of new classes within federated learning (FL). Existing methods often employ generative adversarial networks (GANs) to produce synthetic images to address privacy concerns in FL. However, GANs exhibit inherent instability and high sensitivity, compromising the effectiveness of these methods. In this paper, we introduce a novel data-free federated class incremental learning framework with diffusion-based generative memory (DFedDGM) to mitigate catastrophic forgetting by generating stable, high-quality images through diffusion models. We design a new balanced sampler to help train the diffusion models to alleviate the common non-IID problem in FL, and introduce an entropy-based sample filtering technique from an information theory perspective to enhance the quality of generative samples. Finally, we integrate knowledge distillation with a feature-based regularization term for better knowledge transfer. Our framework does not incur additional communication costs compared to the baseline FedAvg method. Extensive experiments across multiple datasets demonstrate that our method significantly outperforms existing baselines, e.g., over a 4% improvement in average accuracy on the Tiny-ImageNet dataset.

翻译：联邦类增量学习（FCIL）是一个关键但尚未被充分探索的问题，它涉及在联邦学习（FL）中动态纳入新类别。现有方法通常采用生成对抗网络（GAN）来生成合成图像，以应对FL中的隐私担忧。然而，GAN具有固有的不稳定性和高敏感性，这影响了这些方法的有效性。本文提出了一种新颖的、基于扩散生成记忆的无数据联邦类增量学习框架（DFedDGM），通过扩散模型生成稳定、高质量的图像来缓解灾难性遗忘。我们设计了一种新的平衡采样器来帮助训练扩散模型，以缓解FL中常见的非独立同分布问题，并从信息论的角度引入了一种基于熵的样本过滤技术，以提高生成样本的质量。最后，我们将知识蒸馏与基于特征的正则化项相结合，以实现更好的知识迁移。与基准FedAvg方法相比，我们的框架不会产生额外的通信成本。在多个数据集上进行的大量实验表明，我们的方法显著优于现有基线，例如，在Tiny-ImageNet数据集上的平均准确率提升了超过4%。