提出新型去中心化联邦学习协议，实现隐私保护与高效聚合的统一。
FIRMA: FIbonacci Ring Model Aggregation for Privacy-preserving Federated Learning
- 基于斐波那契权重设计环形模型聚合，实现无服务器运行与分类头永久隐私。
- 通过精度门控抑制收敛差的节点，结合环路优化提升全局覆盖与类别多样性。
- 适合高数据异构场景下的隐私敏感联邦学习系统，尤其在标签偏斜下表现卓越。
联邦学习协议面临结构三难困境：传统服务器聚合存在单点故障和梯度反演风险；去中心化环形通信暴露分类头给半诚实同伴；个性化方法重新引入中心化聚合。现有方案无法同时实现无服务器、永久私有分类头、环形拓扑和合理非对称邻居加权。本文提出FIRMA（斐波那契环模型聚合）系列三种逐步增强的联邦学习协议：1） ibfl\ 建立基础：无服务器环形聚合，采用斐波那契权重混合邻居模型，并永久保护分类头隐私；2） ibflp\n引入精度门控的邻居抑制机制，选择性降低收敛差的同伴权重，同时保持斐波那契方向偏好；3） ibflpp\n为完整系统，包含2-opt环路重排以最大化相邻客户端类别多样性，通过$K_g{=}\ ext{lceil} N/2\rceil$次通信传递实现全局环覆盖，以及余弦退火自保留校准。我们证明了收敛速率上界及三项支撑命题，涵盖归一化、覆盖、保留与多样性最优性。在28种配置（4个基准+7种异构性）上的系统实验表明， ibflpp\n在所有12种标签偏斜配置中均优于\fedavg\n，CIFAR-10上$K{=}1$时最高提升达+20.7百分点。在狄利克雷异构条件下， ibflpp\n是所有无服务器协议中的帕累托最优者，在28种配置中17次达到最高准确率。
原文摘要 · Abstract (English)
Federated learning protocols face a structural trilemma: canonical server-based aggregation~\cite{mcmahan2017} creates a single point of failure and gradient inversion risk; decentralised ring-gossip alternatives~\cite{hu2019segmented} expose classification heads to semi-honest peers via uninformed uniform weights; and personalised methods~\cite{collins2021exploiting} reintroduce central aggregation. No existing protocol simultaneously achieves server-free operation, permanently private heads, ring topology, and principled asymmetric neighbour weighting. We propose FIRMA (\textbf{FI}bonacci \textbf{R}ing \textbf{M}odel \textbf{A}ggregation), a family of three progressively enhanced federated learning protocols: 1) \fibfl\ establishes the foundation: server-free ring aggregation with Fibonacci-weighted neighbour blending and permanently private classification heads. 2) \fibflp\ augments this with accuracy-gated neighbour suppression, selectively down-weighting poorly-converged peers while preserving the Fibonacci directional bias. 3) \fibflpp, the full system, completes the family with a 2-opt ring permutation that maximises adjacent-client class diversity, global ring coverage via $K_g{=}\lceil N/2\rceil$ gossip passes, and cosine-annealed self-retention calibration. We establish a convergence rate bound and three supporting propositions governing normalisation, coverage, retention, and diversity optimality. Systematic experiments across 28 configurations -- four benchmarks crossed with seven heterogeneity regimes -- demonstrate that \fibflpp\ surpasses \fedavg\ in all 12 label-skew configurations, with a peak advantage of $+20.7$\,pp on CIFAR-10 at $K{=}1$. Under Dirichlet heterogeneity, \fibflpp\ is the Pareto-dominant method among all server-free protocols, achieving the highest accuracy in 17 of 28 configurations.
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。