跳到正文
arXiv cs.AI· Yang Jiao, Kai Yang, Jianwei Huang·· 4 小时前AI 评分35

FPLL:分布式网络上的联邦软多层学习框架与收敛性保证

Federated Soft Poly-Level Learning Over a Distributed Network with Guaranteed Convergence

AI 导读

研究者提出联邦多层学习框架 FPLL,在分布式网络上以联邦方式求解带软约束的多层学习(PLL)问题,核心是用于逐层松弛的新型非线性割 L-cut。算法提供可调参数,可在计算复杂度与性能之间灵活权衡,并给出达到 ε-Pareto 稳定点的迭代复杂度非渐近分析,实验验证了方法的有效性。该工作已被 IEEE Transactions on Networking 接收。

正文 · 原文

View PDF HTML (experimental)

Abstract:Multi-level learning/optimization, e.g., bilevel optimization, has been recognized for its versatile applicability across a broad array of domains, spanning from signal processing to machine learning. However, most existing works focus on bilevel and trilevel optimization. In contrast, multi-level optimization problems involving four or more hierarchical levels, which we term poly-level learning (PLL) problems, remain largely this http URL addition, in practical PLL, data may be distributed across multiple nodes over a distributed network. Nevertheless, extant studies on PLL predominantly focus on the non-federated setting, necessitating the collection of data to central servers and thereby posing potential privacy risks. To this end, we propose a Federated Poly-Level Learning (FPLL) framework over a distributed network to tackle the PLL problems with soft constraints in a federated manner. FPLL hinges on a novel non-linear cut, i.e., $L$-cut, for the level-wise relaxation to the PLL problems. Additionally, the proposed federated algorithm is designed with adjustable parameters that allow for a flexible trade-off between its computational complexity and performance efficiency, enabling users to tune it according to specific application requirements. We also carry out non-asymptotic analysis to derive the iteration complexity for the proposed FPLL to achieve the $\epsilon$-Pareto stationary point. Extensive experiments have been undertaken to elucidate the superior efficacy of the proposed method.
Comments: Accepted at IEEE Transactions on Networking
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.17722 [cs.AI]
  (or arXiv:2611.17722v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.17722

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yang Jiao [view email]
[v1] Fri, 13 Nov 2026 04:05:12 UTC (102 KB)

来源:arXiv cs.AI · arxiv.org