arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.08615cs.DS
收藏

欧几里得聚类的分布式算法

Distributed Algorithms for Euclidean Clustering

Google Research(谷歌研究院);Texas A&M University(德克萨斯农工大学);Carnegie Mellon University(卡内基梅隆大学)

Vincent Cohen-Addad, Liudeng Wang, David P. Woodruff, Samson Zhou

AI总结:

本文提出分布式环境下欧几里得聚类的高效算法,通过改进通信复杂度实现(1+ε)-近似保证。

摘要:

我们研究了在分布式环境下构造(1+ε)-核集以解决欧几里得(k,z)-聚类问题,其中n个数据点被分成s个站点。我们重点关注两种主要的通信模型:协调模型和黑板模型。在协调模型中,我们设计了一种协议,能够以总通信复杂度为~O(sk + dk/ min(ε⁴, ε^{2+z}) + dk log(nΔ))位的方式实现(1+ε)-强核集,优于先前工作(Chen等,NeurIPS 2016)之处在于消除了在所有服务器之间明文传输显式点坐标的需求。在黑板模型中,我们进一步将通信复杂度降低到~O(s log(nΔ) + dk log(nΔ) + dk/ min(ε⁴, ε^{2+z}))位,优于先前方法,同时将近似保证从常数因子提升到(1+ε)-近似保证。我们的技术结合了新的常数因子近似策略、高效的核集构造和紧凑编码方案,导致最优协议,其通信成本与目前最佳的离线核集构造和现有下界(Chen等,NeurIPS 2016,Huang等,STOC 2024)相匹配,仅相差多项式对数因子。

英文摘要:

We study the problem of constructing $(1+\varepsilon)$-coresets for Euclidean $(k,z)$-clustering in the distributed setting, where $n$ data points are partitioned across $s$ sites. We focus on two prominent communication models: the coordinator model and the blackboard model. In the coordinator model, we design a protocol that achieves a $(1+\varepsilon)$-strong coreset with total communication complexity $\tilde{O}\left(sk + \frac{dk}{\min(\varepsilon^4,\varepsilon^{2+z})} + dk\log(nΔ)\right)$ bits, improving upon prior work (Chen et al., NeurIPS 2016) by eliminating the need to communicate explicit point coordinates in-the-clear across all servers. In the blackboard model, we further reduce the communication complexity to $\tilde{O}\left(s\log(nΔ) + dk\log(nΔ) + \frac{dk}{\min(\varepsilon^4,\varepsilon^{2+z})}\right)$ bits, achieving better bounds than previous approaches while upgrading from constant-factor to $(1+\varepsilon)$-approximation guarantees. Our techniques combine new strategies for constant-factor approximation with efficient coreset constructions and compact encoding schemes, leading to optimal protocols that match both the communication costs of the best-known offline coreset constructions and existing lower bounds (Chen et al., NeurIPS 2016, Huang et. al., STOC 2024), up to polylogarithmic factors.

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。

↑