CRIP：面向个性化单轮联邦学习的通道级表示注入

中文摘要 AI 辅助

单轮联邦学习（OSFL）作为一种仅需一轮通信的协作模型学习框架，已展现出良好的应用前景，在通信效率和隐私保护方面具有显著优势。然而，由于客户端间存在严重的域异质性，且缺乏迭代知识交换，OSFL常面临固有局限。现有多数OSFL方法需借助辅助公共数据集进行知识蒸馏，或利用统计信息开展参数级聚合，却忽略了域异质性导致的特征偏移。为应对这些挑战，本文提出CRIP，这是一种在表示空间中通过通道级特征对齐运行的个性化OSFL框架。为实现该目标，每个客户端将其特征提取器上传至服务器，服务器再将所有提取器广播回每个客户端。由于并非所有源客户端都与目标客户端共享兼容的特征分布，不加区分地融合跨客户端特征会引入域特定噪声。因此，CRIP在小型本地小批量数据上有效测量目标客户端与每个源客户端间的通道级表示相似度，仅选择性融合最兼容的特征。在DomainNet、PACS和Office-Home等域异基准上开展的大量实验表明，CRIP始终优于本地模型和最先进的基线方法，验证了在极端域异质性下表示空间个性化的有效性。

英文摘要

One-shot federated learning (OSFL) has emerged as a promising collaborative model learning framework with only a single round of communication, offering significant advantages in communication efficiency and privacy preservation. However, OSFL often faces inherent limitations under severe domain heterogeneity across clients due to the lack of iterative knowledge exchange. Most existing OSFL methods require an auxiliary public dataset for knowledge distillation or leverage statistical information for parameter-level aggregation, overlooking feature shift caused by domain heterogeneity. To address these challenges, we propose CRIP, a personalized OSFL framework that operates in the representation space via channel-level feature alignment. To achieve this, each client uploads its feature extractor to the server, which broadcasts all extractors back to every client. Since not all source clients share compatible feature distributions with the target client, indiscriminate fusion of cross-client features would introduce domain-specific noise. Therefore, CRIP effectively measures the channel-wise representational similarity between the target client and each source client on a small local mini-batch, and selectively fuses only the most compatible features. Extensive experiments on domain-heterogeneous benchmarks such as DomainNet, PACS, and Office-Home demonstrate that CRIP consistently outperforms local models and state-of-the-art baselines, validating the effectiveness of representation-space personalization under extreme domain heterogeneity.

发表机构

• University of Surrey（萨里大学）
• Taiyuan University of Science and Technology（太原科技大学）
• Xidian University（西安电子科技大学）

机构由 AI 辅助整理，请以论文原文为准。