克服联邦类增量学习中的灾难性遗忘：基于联邦全局孪生生成器
Overcoming Catastrophic Forgetting in Federated Class-Incremental Learning via Federated Global Twin Generator
AI总结：
提出联邦全局孪生生成器（FedGTG），在服务器端训练生成器合成数据，客户端用特征方向控制损失，缓解联邦类增量学习中的灾难性遗忘，提升准确率并降低遗忘。
摘要：
联邦类增量学习（FCIL）在去中心化环境中日益重要，它使多个参与者能够在不共享私有数据的情况下协作训练一个全局模型，以在一系列任务上表现良好。在FCIL中，传统的联邦学习算法（如FedAVG）常常遭受灾难性遗忘，导致在早期任务上性能显著下降。最近基于生成模型的工作通过生成合成图像来帮助缓解所有类别上的这一问题，但这些方法在先前类别上的测试准确率仍远低于近期类别，即具有更好的可塑性而非稳定性。为克服这些问题，本文提出了联邦全局孪生生成器（FedGTG），一种FCIL框架，它在全局侧利用隐私保护的生成模型训练，而无需访问客户端数据。具体而言，服务器训练一个数据生成器和一个特征生成器，以从所有已见类别中创建两类信息，然后将合成数据发送到客户端。客户端随后使用特征方向控制损失，使本地模型保留知识并良好地学习新任务。我们广泛分析了FedGTG在自然图像上的鲁棒性，以及其收敛到平坦局部最小值并实现更好预测置信度（校准）的能力。在CIFAR-10、CIFAR-100和tiny-ImageNet上的实验结果表明，与先前框架相比，FedGTG在准确性和遗忘度量方面均有改进。
英文摘要：
Federated Class-Incremental Learning (FCIL) increasingly becomes important in the decentralized setting, where it enables multiple participants to collaboratively train a global model to perform well on a sequence of tasks without sharing their private data. In FCIL, conventional Federated Learning algorithms such as FedAVG often suffer from catastrophic forgetting, resulting in significant performance declines on earlier tasks. Recent works, based on generative models, produce synthetic images to help mitigate this issue across all classes, but these approaches' testing accuracy on previous classes is still much lower than recent classes, i.e., having better plasticity than stability. To overcome these issues, this paper presents Federated Global Twin Generator (FedGTG), an FCIL framework that exploits privacy-preserving generative-model training on the global side without accessing client data. Specifically, the server trains a data generator and a feature generator to create two types of information from all seen classes, and then it sends the synthetic data to the client side. The clients then use feature-direction-controlling losses to make the local models retain knowledge and learn new tasks well. We extensively analyze the robustness of FedGTG on natural images, as well as its ability to converge to flat local minima and achieve better-predicting confidence (calibration). Experimental results on CIFAR-10, CIFAR-100, and tiny-ImageNet demonstrate the improvements in accuracy and forgetting measures of FedGTG compared to previous frameworks.
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。
