English

Fully Dynamic $k$-Clustering with Fast Update Time and Small Recourse

Data Structures and Algorithms 2024-08-05 v1

Abstract

In the dynamic metric kk-median problem, we wish to maintain a set of kk centers SVS \subseteq V in an input metric space (V,d)(V, d) that gets updated via point insertions/deletions, so as to minimize the objective xVminySd(x,y)\sum_{x \in V} \min_{y \in S} d(x, y). The quality of a dynamic algorithm is measured in terms of its approximation ratio, "recourse" (the number of changes in SS per update) and "update time" (the time it takes to handle an update). The ultimate goal in this line of research is to obtain a dynamic O(1)O(1) approximation algorithm with O~(1)\tilde{O}(1) recourse and O~(k)\tilde{O}(k) update time. Dynamic kk-median is a canonical example of a class of problems known as dynamic kk-clustering, that has received significant attention in recent years. To the best of our knowledge, however, previous papers either attempt to minimize the algorithm's recourse while ignoring its update time, or minimize the algorithm's update time while ignoring its recourse. For dynamic kk-median, we come arbitrarily close to resolving the main open question on this topic, with the following results. (I) We develop a new framework of randomized local search that is suitable for adaptation in a dynamic setting. For every ϵ>0\epsilon > 0, this gives us a dynamic kk-median algorithm with O(1/ϵ)O(1/\epsilon) approximation ratio, O~(kϵ)\tilde{O}(k^{\epsilon}) recourse and O~(k1+ϵ)\tilde{O}(k^{1+\epsilon}) update time. This framework also generalizes to dynamic kk-clustering with p\ell^p-norm objectives, giving similar bounds for the dynamic kk-means and a new trade-off for dynamic kk-center. (II) If it suffices to maintain only an estimate of the value of the optimal kk-median objective, then we obtain a O(1)O(1) approximation algorithm with O~(k)\tilde{O}(k) update time. We achieve this result via adapting the Lagrangian Relaxation framework to the dynamic setting.

Keywords

Cite

@article{arxiv.2408.01325,
  title  = {Fully Dynamic $k$-Clustering with Fast Update Time and Small Recourse},
  author = {Sayan Bhattacharya and Martín Costa and Naveen Garg and Silvio Lattanzi and Nikos Parotsidis},
  journal= {arXiv preprint arXiv:2408.01325},
  year   = {2024}
}

Comments

Accepted at FOCS 2024

R2 v1 2026-06-28T18:02:22.990Z