arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.14830math.OCcs.SYeess.SPeess.SY
收藏

关于Network-GIANT的收敛性分析:一种基于近似海森矩阵的全分布式优化算法

On Convergence Analysis of Network-GIANT: An approximate Hessian-based fully distributed optimization algorithm

Uppsala University(乌普萨拉大学);University of Padova(帕多瓦大学)

Souvik Das, Luca Schenato, Subhrakanti Dey

AI总结:

研究针对\(L\)-光滑、\(\mu\)-强凸局部损失函数的近似牛顿型全分布式优化方法Network-GIANT,明确其全局线性收敛率,给出步长参数界限及最优性差距范数上界,通过实验验证理论结果。

摘要:

本文对一种新开发的近似牛顿型全分布式优化方法Network-GIANT进行了详细的收敛性和性能分析,该方法用于\(L\)-光滑、\(\mu\)-强凸局部损失函数。明确了其全局线性收敛率,给出步长参数界限,推导了最优性差距范数的混合线性-二次不等式上界,并通过数值实验验证理论结果。

英文摘要:

This paper presents a detailed convergence and performance analysis of a recently developed approximate Newton-type fully distributed optimization method for \(L\)-smooth, \(μ\)-strongly convex local loss functions, called Network-GIANT (inspired by the Federated learning algorithm GIANT possessing mixed linear-quadratic convergence properties). Network-GIANT has been empirically seen to achieve faster linear convergence properties compared to its gradient-based counterparts, and several other existing second order distributed algorithms, while having the same communication complexity (per iteration) as its first order distributed counterparts. We first explicitly characterize a \emph{global linear convergence rate} for Network-GIANT, which can be computed as the spectral radius of a $3 \times 3$ matrix dependent on $L$, $μ$, and the spectral norm ($σ$) of the consensus matrix of the underlying undirected graph. We provide an explicit bound on the step size parameter $η$, below which this spectral radius is guaranteed to be less than $1$. Furthermore, we derive a mixed linear-quadratic inequality based upper bound for the optimality gap norm, and provide a rigorous proof of a local asymptotic convergence rate of \(1 - η\big(1 - \fracγμ\big)\) given the Hessian approximation error $γ< μ$, which formally explains the faster convergence rate of Network-GIANT. Numerical experiments are carried out with a reduced CovType dataset for binary logistic regression over a variety of graphs, including heterogeneous data distributions, to illustrate the above theoretical results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑