Searcharxiv⌕ Search

arXiv · 2611.14573

Beyond the First Hit: Towards Tight Guarantees for $k$-means++

Abstract

Let $R_k$ be the worst-case ratio between the expected cost of the $k$ seeds chosen by $k$-means++ and the optimal $k$-means cost. In their seminal paper introducing $k$-means++, Arthur and Vassilvitskii (SODA 2007) showed that $2\ln k\le R_k\le8(\ln k+2)$, and Makarychev, Reddy and Shan (NeurIPS 2020) improved the upper bound to $5(\ln k+2)$. We prove $R_k\le\frac{1163}{233}+2.108\,H_{k-2}$ for every $k\ge3$ by a computer-assisted argument, and $R_k\le5+2.3139\,H_{k-2}$ for every $k\ge2$ by a short analytic argument; here $H_j$ is the $j$th harmonic number. Since $H_{k-2}\le1+\ln k$, this lowers the leading coefficient of $\ln k$ from $5$ to $2.108$. We also determine $R_2=5$ and $R_3=185/32$ exactly; both are suprema that are approached but not attained. On the lower side we prove $R_k\ge2(1+H_{k-1})$ for every $k$, and give a second family of lower bounds that is larger than this for $2\le k\le18$. Previous analyses bound the expected cost of a cluster just after it is first sampled by a constant times its optimal cost, and then charge that cost for the rest of the run. We instead track how a cluster's cost evolves after its first hit: a cluster left with a large residual cost is likely to be sampled again before it accumulates much of the logarithmic charge, and each repeat repairs it.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Aravind Reddy. 2026-11-09. Beyond the First Hit: Towards Tight Guarantees for $k$-means++. https://searcharxiv.com/abs/2611.14573

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

A Black-Box Approach for Exogenous Replenishment in Online Resource Allocation

In a typical online resource allocation problem, we start with a fixed inventory of resources and make online allocation decisions in response to resource requests that arrive sequentially over a finite horizon. We consider settings where the inventory is replenished over time according to an unknown exogenous process. We introduce black-box methods that extend any existing algorithm, originally designed without considering replenishment, into one that works with an arbitrary (adversarial or stochastic) replenishment process. Our approach preserves the original algorithm's competitive ratio in regimes with large initial inventory, thereby enabling the seamless integration of exogenous replenishment into a large body of existing algorithmic results for both adversarial and stochastic arrival models.

cs.DS↗

PHastR: Faster Minimal Perfect Hashing with Rotated Layouts

We present PHastR, a minimal perfect hash function of the PHast family. Like PHast+, the fast-to-build variant of PHast, PHastR finds feasible seeds with bit-parallel operations: its seeds move keys along rings of slots, and to evaluate the seeds of a bucket it cyclically shifts, for each key, a word recording which slots of its ring are used. Unlike PHast+, however, it is almost immune to self-collisions, which make two keys of a bucket collide for (almost) all its seeds: more than a third of the buckets bumped by PHast+ with wrapping and multiplier 3 contain one. With 8-bit seeds, PHastR uses $1.94$ bits per key, essentially the same space as the best variant of PHast+ with wrapping, but it is built in $37\%$ of the time, and its queries are $34\%$ faster on $10^9$ keys, as it bumps $0.5\%$ of the keys instead of $3.9\%$; they are faster also than those of PHast, which takes almost thirteen times as long to build, and than those of PtrHash, which uses more than one bit per key more. With a larger expected bucket size PHastR comes within $0.02$ bits per key of the space of PHast, using $1.90$ bits per key, and its queries are still about as fast. Its construction also scales better with the number of threads: with all the threads of a 96-core machine it is more than two and a half times faster than that of PHast+. An offline construction built PHastR on a trillion keys in $9.3$ hours on a single machine. Finally, we argue that when each key can be placed only within a range of $W$ slots, every placement of $n$ random keys on $n$ slots, however found, must bump on large key sets a constant fraction of the keys: we derive an asymptotic lower bound of about $0.1/W$ on the expected value of this fraction, and we prove that, in a Poisson model, the best placement bumps about $1/2W$ of the keys, which is about one tenth of the keys bumped by PHastR.

cs.DS↗

Single-Vertex Metropolis on Minor-Free Graphs: A General EPTAS

We consider a class of combinatorial optimization problems where the goal is to find a largest induced subgraph with a prescribed property. We assume that the property is preserved by taking induced subgraphs and disjoint unions, holds for all edgeless graphs, and can be tested in polynomial time. We study the single-vertex Metropolis process for this class of problems. We show that on graphs excluding an arbitrary fixed minor $H$, this process is a randomized efficient polynomial-time approximation scheme. Starting from any feasible solution on a graph with $n$ vertices, it finds a $(1-\varepsilon)$-approximation within $\exp(O_H(\varepsilon^{-2})) n(n+\log(1/δ))$ steps with probability at least $1-δ$. This includes maximum independent set, maximum induced forest, maximum induced bipartite subgraph, and maximum induced subgraph of degree at most one. The proof uses improving exchanges on one fixed division of the input graph into regions. A recursive separator construction performs each exchange inside a region of size $r$ in at most $2r$ moves, with temporary loss $O_H(\sqrt r)$. We count move sequences through each transition and bound the probability of avoiding all near-optimal states. The general theorem bounds the number of Metropolis steps in terms of the temporary loss along improvement paths. We extend the result to bounded weight imbalance and give a weighted-star obstruction for unrestricted weights.

cs.DS↗