FedARA：异构边缘设备上基于锚驱动表征对齐的资源自适应低秩个性化联邦学习
FedARA: Resource-adaptive Low-rank Personalized Federated Learning via Anchor-driven Representation Alignment on Heterogeneous Edge Devices
低秩分解+重构融合的共享特征提取器方案，允许各客户端按本地资源自选秩值，可直接迁移到视觉模型在异构边缘设备上的分布式协同训练与通信压缩；锚驱动的表征一致性学习则为解决非IID视觉数据下特征空间漂移提供了可复用的对齐思路。
原文依据 we design a low-rank decomposition and reconstruction fusion scheme for shared feature extractors based on the matrix decomposition technology
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论个性化联邦学习（PFL）因能使参与客户端在非独立同分布（non-IID）本地数据上训练定制的个性化模型而受到广泛关注。然而，现有PFL方法主要存在两点局限：1）仅个性化部分支持异构设计，而共享部分必须保持同构；2）在具有非IID数据特性的不同客户端上生成的模型，其语义表征不可避免地趋于不一致，从而对模型性能产生负面影响。为解决上述问题，本文提出一种新颖的、基于锚驱动表征对齐的资源自适应个性化联邦学习框架（FedARA）。具体而言，我们基于矩阵分解技术，为共享特征提取器设计了一种低秩分解与重构融合方案，其中每个客户端可根据其本地可用资源自主设定秩值，从而控制提取器的复杂度，并自然地降低通信与计算开销。此外，针对跨客户端特征空间不一致的问题，我们提出了一种锚驱动的表征一致性学习机制，该机制能够引导客户端模型学习统一的特征表征，并缓解全局知识遗忘，从而提升个性化模型的性能。大量实验结果表明，在多种异构场景下，我们的方法以更低的通信与计算开销显著优于十七个最优基线方法。
英文原文摘要 1547 字符
Personalized Federated Learning (PFL) has gained significant attention for enabling participating clients to train customized personalized models on non-IID local data. However, current PFL methods mainly suffer from two limitations: 1) Only the personalized part supports heterogeneous design, while the shared part must remain homogeneous. 2) The semantic representations of models generated on different clients with non-IID data characteristics inevitably tend to be inconsistent, negatively impacting model performance. To conquer them, this paper proposes a novel resource-adaptive personalized Federated Learning via Anchor-driven Representation Alignment (FedARA). Concretely, we design a low-rank decomposition and reconstruction fusion scheme for shared feature extractors based on the matrix decomposition technology, where each client can autonomously set the rank value based on its locally available resources, controlling the complexity of extractors and naturally reducing communication and computational costs. Moreover, to address the inconsistency of feature spaces across clients, an anchor-driven representation consistency learning mechanism is developed, which can guide client models to learn unified feature representations and alleviate global knowledge forgetting, thereby improving personalized model performance. Extensive experimental results demonstrate that our method significantly outperforms seventeen state-of-the-art baselines in diverse heterogeneous scenarios with less communication and computational costs.