提出轻量级联邦持续学习算法，解决数据分布漂移下的知识遗忘问题。
Server-Proximal Aggregation for Federated Domain-Incremental Learning under Partial Participation: Task-Uniform Convergence and Backward Transfer
- 服务器添加锚点项，引导客户端更新向历史模型靠近，防止知识漂移。
- 理论证明可控制先验任务损失增长，且收敛速率达非凸最优水平。
- 无需记忆缓冲或任务头，适合隐私敏感的分布式场景，尤其适合资源受限设备。
现实中的联邦系统很少在静态数据上运行：输入分布随时间漂移，而隐私规则禁止原始数据共享。本文研究此场景下的联邦域增量学习（FDIL），其中（i）客户端异构，（ii）任务按顺序到来且领域发生漂移，但（iii）标签空间保持不变。现有方法在真实部署下仍缺乏两个关键理论保障：后向知识迁移（BKT）的保证，以及在部分参与条件下对所有任务序列的收敛速率。本文提出SPECIAL（Server-Proximal Efficient Continual Aggregation for Learning），一种简单、无内存的FDIL算法，仅在标准FedAvg基础上增加一个服务器端的“锚点”：每轮中，服务器通过轻量级近端项将随机抽样的参与客户端更新向先前全局模型微调。该锚点在不使用回放缓冲、合成数据或任务特定头的前提下抑制累积漂移，同时保持通信和模型规模不变。理论分析表明，SPECIAL（i）保留早期任务性能：先验任务损失的增长被一个受漂移控制的项约束，且随轮数、本地迭代次数和参与客户端数增加而减小；（ii）高效学习所有任务：首次获得在部分参与下的非凸收敛率，为O((E/NT)^(1/2))，其中E为本地迭代次数，T为通信轮数，N为每轮参与客户端数，与单任务FedAvg匹配，并显式分离优化方差与跨任务漂移。实验进一步验证了SPECIAL的有效性。
原文摘要 · Abstract (English)
Real-world federated systems seldom operate on static data: input distributions drift while privacy rules forbid raw-data sharing. We study this setting as Federated Domain-Incremental Learning (FDIL), where (i) clients are heterogeneous, (ii) tasks arrive sequentially with shifting domains, yet (iii) the label space remains fixed. Two theoretical pillars remain missing for FDIL under realistic deployment: a guarantee of backward knowledge transfer (BKT) and a convergence rate that holds across the sequence of all tasks with partial participation. We introduce SPECIAL (Server-Proximal Efficient Continual Aggregation for Learning), a simple, memory-free FDIL algorithm that adds a single server-side ``anchor'' to vanilla FedAvg: in each round, the server nudges the uniformly sampled participated clients update toward the previous global model with a lightweight proximal term. This anchor curbs cumulative drift without replay buffers, synthetic data, or task-specific heads, keeping communication and model size unchanged. Our theory shows that SPECIAL (i) preserves earlier tasks: a BKT bound caps any increase in prior-task loss by a drift-controlled term that shrinks with more rounds, local epochs, and participating clients; and (ii) learns efficiently across all tasks: the first communication-efficient non-convex convergence rate for FDIL with partial participation, O((E/NT)^(1/2)), with E local epochs, T communication rounds, and N participated clients per round, matching single-task FedAvg while explicitly separating optimization variance from inter-task drift. Experimental results further demonstrate the effectiveness of SPECIAL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。