English

Space Complexity of Euclidean Clustering

Computational Geometry 2025-03-18 v3 Data Structures and Algorithms

Abstract

The (k,z)(k, z)-Clustering problem in Euclidean space Rd\mathbb{R}^d has been extensively studied. Given the scale of data involved, compression methods for the Euclidean (k,z)(k, z)-Clustering problem, such as data compression and dimension reduction, have received significant attention in the literature. However, the space complexity of the clustering problem, specifically, the number of bits required to compress the cost function within a multiplicative error ε\varepsilon, remains unclear in existing literature. This paper initiates the study of space complexity for Euclidean (k,z)(k, z)-Clustering and offers both upper and lower bounds. Our space bounds are nearly tight when kk is constant, indicating that storing a coreset, a well-known data compression approach, serves as the optimal compression scheme. Furthermore, our lower bound result for (k,z)(k, z)-Clustering establishes a tight space bound of Θ(nd)\Theta( n d ) for terminal embedding, where nn represents the dataset size. Our technical approach leverages new geometric insights for principal angles and discrepancy methods, which may hold independent interest.

Keywords

Cite

@article{arxiv.2403.02971,
  title  = {Space Complexity of Euclidean Clustering},
  author = {Xiaoyi Zhu and Yuxiang Tian and Lingxiao Huang and Zengfeng Huang},
  journal= {arXiv preprint arXiv:2403.02971},
  year   = {2025}
}

Comments

Accepted by SoCG2024, TIT2025, in IEEE Transactions on Information Theory, 2025

R2 v1 2026-06-28T15:09:47.971Z