在持续联邦学习中使用扩散模型作为生成式回放——将会发生什么？
Using Diffusion Models as Generative Replay in Continual Federated Learning -- What will Happen?
AI总结：
针对持续联邦学习中的灾难性遗忘与非独立同分布数据挑战，本文提出DCFL框架，利用条件扩散模型在本地设备生成合成历史数据进行生成式回放，有效缓解数据分布偏移，并提供了收敛界及多数据集验证。
摘要：
联邦学习（FL）已成为去中心化学习的基石，在许多场景中，输入数据分布会随时间动态变化，从而引入持续学习（CL）问题。这种持续联邦学习（CFL）任务带来了独特的挑战，尤其是在灾难性遗忘和非独立同分布（non-IID）输入数据方面。现有的解决方案包括使用回放缓冲区来存储历史数据，或利用生成对抗网络。然而，受扩散模型在生成任务中最新进展的启发，本文提出了DCFL，这是一种专为解决动态分布式学习环境中CFL挑战而定制的新型框架。该方法利用条件扩散模型的能力，在通信期间于每个本地设备上生成合成历史数据，有效缓解了动态数据分布输入中潜在的偏移。我们为所提出的CFL框架提供了收敛界，并在多个数据集上展示了其优异的性能，证明了其在应对CFL任务复杂性方面的有效性。
英文摘要：
Federated learning (FL) has become a cornerstone in decentralized learning, where, in many scenarios, the incoming data distribution will change dynamically over time, introducing continuous learning (CL) problems. This continual federated learning (CFL) task presents unique challenges, particularly regarding catastrophic forgetting and non-IID input data. Existing solutions include using a replay buffer to store historical data or leveraging generative adversarial networks. Nevertheless, motivated by recent advancements in the diffusion model for generative tasks, this paper introduces DCFL, a novel framework tailored to address the challenges of CFL in dynamic distributed learning environments. Our approach harnesses the power of the conditional diffusion model to generate synthetic historical data at each local device during communication, effectively mitigating latent shifts in dynamic data distribution inputs. We provide the convergence bound for the proposed CFL framework and demonstrate its promising performance across multiple datasets, showcasing its effectiveness in tackling the complexities of CFL tasks.
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。
