FedHarmony：协调联邦多标签学习中的异构标签相关性
FedHarmony: Harmonizing Heterogeneous Label Correlations in Federated Multi-Label Learning
对做多标签视觉识别（如多属性识别、多标签分类）的研究者，这篇论文提供了在数据分散、标签空间不一致时校正标签共现结构漂移的思路：用跨客户端的共识相关性作为全局教师来纠正本地偏置估计，并可复用于任何需要建模标签共现的分布式或去中心化训练场景。
原文依据 correlations learned by individual clients inevitably deviate from the global structure, a phenomenon we term label correlation drift
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论联邦多标签学习是一种分布式范式，其中多个客户端持有异构的多标签数据，并在不共享原始数据的隐私约束下进行协同学习。然而，在异构分布下建模标签相关性仍然具有挑战性。由于各客户端特有的标签空间以及各不相同的共现模式，单个客户端学习到的相关性不可避免地偏离全局结构，我们将这一现象称为标签相关性漂移。为解决该问题，我们提出 FedHarmony，一个协调各客户端之间异构标签相关性的框架。它引入共识相关性，捕捉客户端之间的一致性，并作为全局教师来纠正有偏的本地估计。在聚合过程中，FedHarmony 同时依据数据规模和相关性质量评估每个客户端，并据此分配权重。此外，我们开发了一种加速优化算法，并在理论上证明了在不牺牲精度的前提下实现更快的收敛。在真实世界的联邦多标签数据集上的实验表明，FedHarmony 持续优于最优方法。
英文原文摘要 1151 字符
Federated Multi-Label Learning is a distributed paradigm where multiple clients possess heterogeneous multi-label data and perform collaborative learning under privacy constraints without sharing raw data. However, modeling label correlations under heterogeneous distributions remains challenging. Due to client-specific label spaces and varying co-occurrence patterns, correlations learned by individual clients inevitably deviate from the global structure, a phenomenon we term label correlation drift. To address this, we propose FedHarmony, a framework that harmonizes heterogeneous label correlations across clients. It introduces consensus correlation, capturing agreement among clients and serving as a global teacher to correct biased local estimates. During aggregation, FedHarmony evaluates each client by both data size and correlation quality, assigning weights accordingly. Moreover, we develop an accelerated optimization algorithm and theoretically establish faster convergence without sacrificing accuracy. Experiments on real-world federated multi-label datasets show that FedHarmony consistently outperforms state-of-the-art methods.