CS Paper Radar

Constant-Round Massively Parallel Algorithms for Approximate MSTs in Metric Spaces

Eunjin Oh, Hyeonjun Shin

cs.DS cs.CGcs.DC ·submitted 2026-11-04 07:35 UTC · v1
theory

Abstract

We develop a unified algorithmic framework for computing approximate minimum spanning trees (MSTs) in metric spaces in the fully scalable Massively Parallel Computation (MPC) model. Using this framework, we obtain the following results for any fixed constant $σ\in(0,1)$: - General metrics: Given all pairwise distances, we compute a $(1+\varepsilon)$-approximate MST in $O(\log(1/\varepsilon))$ rounds using $n^σ$ local space per machine and $\widetilde{O}(n^2)$ total space. - Euclidean spaces: Given $n$ points in $\mathbb{R}^d$, we compute an $O(1/σ^4)$-approximate MST in $O(1)$ rounds using $(nd)^σ$ local space per machine and $O(nd+n^{1+σ})$ total space. - Doubling metrics: Given $n$ points in a metric of doubling dimension $\mathrm{ddim}=O(σ\log n)$, we compute an $O(1/σ)$-approximate MST in $O(\mathrm{ddim}/σ^4)$ rounds using $n^σ$ local space per machine and $\widetilde{O}(n^{1+σ})\cdot 2^{O(\mathrm{ddim})}$ total space. For any fixed constant $\varepsilon\in(0,1)$, these results give the first constant-round MPC algorithms achieving $(1+\varepsilon)$-approximation in general metrics and $O(1)$-approximation in a high-dimensional Euclidean space, respectively. They improve upon the $O(\log\log n+\log(1/\varepsilon))$-round algorithm for general metrics [SODA'25] and the $O(\log\log n\cdot\log\log\log n)$-round algorithm for Euclidean spaces [SODA'24], respectively. Moreover, our round complexity for general metrics is asymptotically optimal in its dependence on $\varepsilon$, matching the $Ω(\log(1/\varepsilon))$ lower bound of [SODA'25] under the 1-vs-2-cycles conjecture. Our result also answers an open question raised by [STOC'23].

Closest papers

Nearest neighbours by meaning across the whole index.

Fast Geometric Spanners via Approximate Nearest Neighbor Search

cs.DS cs.CG ·2026-09-22 ·Alexandr Andoni, Manuel Paez, Krish Singal, Tian Zhang
similarity 0.70
Abstract

We study the problem of constructing metric spanners in general metric spaces in subquadratic time when given blackbox access to a fast algorithm for batch approximate nearest neighbor search. In particular, we show the following results for any metric space $\mathsf{M} = ([n], \mathsf{d})$ with aspect ratio $Δ$ admitting a $c$-approximate batch nearest neighbor search algorithm with runtime $τ_{\mathsf{M}}(n)$, (1) There exists an algorithm that, for any $k \in \mathbb{N}$, constructs an $O(c k)$-distortion spanner with $\tilde{O}(kn^{1+1/2k} \log Δ)$ edges and runs in time $\tilde{O}(τ_{\mathsf{M}} \cdot k n^{1/k} \log Δ)$. (2) Any algorithm that learns at most $o(n^{1+1/k}/k)$ pairwise distances by querying a distance oracle and a blackbox batch nearest neighbor search oracle necessarily incurs $Ω(c k)$ distortion. Our results entail that (truly) sub-quadratic time algorithms for spanner construction is equivalent to subquadratic time BANN (up to constant-factor losses). As a further application, we use our fast spanner constructions to obtain a fast algorithm for approximating the Wasserstein distance $\mathsf{W}_q$, for all $q > 1$, over any metric space admitting an efficient batch approximate nearest neighbor search algorithm. Together with recent new efficient algorithms for approximate nearest neighbor search in $\ell_p$ spaces, for $p > 2$, our results entail the first subquadratic time algorithms for spanner construction (with the stated size-distortion tradeoff) and $\mathsf{W}_q$ distance approximation over these metric spaces.

0.07

Geometric Hashing in MPC with Applications to Clustering and Stochastic Embeddings

cs.DS cs.CG ·2026-10-15 ·Sujoy Bhore, Arnold Filtser, Eunjin Oh, Hyeonjun Shin, Himanshi Singh
top venue similarity 0.69
Abstract · To appear in SODA'27

In this paper we study consistent hashing (a.k.a. sparse partitions) for doubling metrics in the Massively Parallel Computation (MPC) model. A $(Γ,Λ)$-hash with diameter bound $Δ$ partitions a metric space into clusters of diameter at most $Δ$ such that every ball of radius $Δ/Γ$ intersects at most $Λ$ clusters in expectation. Consistent hashing has emerged as a central tool in geometric MPC algorithms. However, so far it has been solely used in Euclidean spaces, where it can be pre-computed. On the other hand, in cases where the metric space is received as input and unknown in advance, no previous parallel construction was known. We present the first fully scalable MPC algorithm over machines with local space $n^σ$, that given an $n$-point metric space with doubling dimension $d\leq O(σ\log n)$, samples a randomized $(Γ,2^{O(d/Γ)})$-hash in $O(d/σ^3)$ rounds. There is a dichotomy: if the doubling dimension is $d=Ω(σ\cdot\log n)$, a polynomial number of rounds is required to compute such a hash. We conclude by showing that this MPC primitive yields $O(d/σ^3)$-round fully scalable MPC algorithms for a broad class of fundamental geometric problems in doubling metrics, including stochastic tree embeddings with $O(\log n)$ expected distortion, constant factor approximation for facility location, bicriteria $(O(ε^{-2}),1+ε)$-approximation for $k$-median and $k$-means.

0.73

How to Walk a Dog in Parallel: On Parallel Computation of the Discrete Frechet Distance

cs.DS cs.CG ·2026-10-21 ·Leonid Barenboim, Arnold Filtser, Omrit Filtser, Orr Fischer
similarity 0.66
Abstract

The Fréchet distance (also known as the dog-leash distance) is a well-studied and prevalent distance measure for polygonal curves. In this paper we present parallel algorithms for computing the discrete \frechet\ distance between two curves, in both the PRAM and MPC models. For two polygonal curves of length $n$, we show that the distance between them can be computed in the PRAM model in $O(\log n/ σ)$ rounds and $\tilde{O}(n^{2+σ})$ total work for any $0<σ<1$. In the sublinear MPC model with $n^ρ$ space per machine, it can be computed in $O(\log n / ρ^2)$ rounds and $\tilde{O}(n^{2+ρ})$ total space. For $c$-packed curves, we match our bounds to the sequential algorithms, showing that a $(1+\eps)$-approximation of the distance can be computed in the PRAM model in $O(\log{n}+1/σ^3)$ rounds and $\tilde{O}(n^{1+σ})$ total work, and in the sublinear MPC model in $O(1 /ρ^3)$ rounds and $\tilde{O}(n^{1+ρ})$ total space, for any constant $\eps>0$. As part of our algorithms for $c$-packed curves, we also present parallel algorithms for computing curve simplifications, which may be of independent interest.

0.17

Dimension Thresholds for Relative-Error Representations of Finite Metrics in Pseudo-Euclidean Spaces

cs.CG math.MG ·2026-10-16 ·Nikita Muravya, Aleksandr Bespalov, Anna Vozianova
theory similarity 0.65
Abstract

A finite metric space with $N$ points admits an exact representation in a pseudo-Euclidean space of dimension at most $N-1$. We study the minimum dimension required for relative-error approximation of squared distances. Our main result gives an upper bound for arbitrary metrics in terms of local distance ratios: for $4/5<\varepsilon<1$, dimension $O((N/k)\log N+kρ_k^4\log N/(\varepsilon-4/5)^2)$ suffices, where $ρ_k$ is the largest ratio, over all points, between the distance to the $k$th closest point (counting the point itself) and the nearest-neighbour distance. The proof combines a low-rank landmark approximation with a sparse correction controlled by its factorization norm. For metrics with distances in $[1,Φ]$, where $Φ\ge5$ is fixed, the worst-case dimension is $Θ(N)$ for $0<\varepsilon<4/5$, lies between $Ω(\sqrt N)$ and $O(\sqrt N\log N)$ for $4/5<\varepsilon<12/13$, and is $Θ(\log N)$ for $\varepsilon>(Φ^2-1)/(Φ^2+1)$. For $Φ>5$, an intermediate range remains without matching bounds. Sign-pattern counting yields the $Ω(\sqrt N)$ lower bound for $4/5<\varepsilon<12/13$, while Hadamard matrices give explicit linear lower bounds for $\varepsilon<4/5$ via a known approximate-rank estimate. We also construct configurations with null squared intervals whose minimum representation dimension remains $N-1$ for every $\varepsilon<1$. Finally, for every fixed $0<\varepsilon<3/5$, we show that the quadratic dependence on the parameter $C$ in the dimension bound derived from Deng et al.'s pseudo-Euclidean Johnson-Lindenstrauss lemma is necessary up to a logarithmic factor.

0.22

On the Metric Distortion of the Minimum Spanning Tree Problem

cs.DS ·2026-10-16 ·Kian Ghassemi, Masoud Seddighin, Saeed Seddighin
similarity 0.65
Abstract · 24 pages, 9 figures, 1 table

We study the metric distortion of the minimum spanning tree (MST) problem when numerical edge weights are unavailable and each vertex reports only a ranking of the other vertices induced by these weights. We first show, through a simple example, that without assumptions on the weights, no ordinal algorithm guarantees a finite approximation ratio. We therefore focus on general metrics and present a Borůvka-inspired algorithm using only ordinal comparisons that achieves distortion $O(\log n)$. We complement this result by showing that, for every $\varepsilon>0$, no deterministic ordinal algorithm guarantees distortion at most $3/2-\varepsilon$ on all finite metric instances. We then investigate more informative settings in pursuit of constant distortion, first considering two well-studied metric classes. For tree metrics realized on the given vertex set, a more careful variant of our initial algorithm achieves distortion~$1$, recovering an optimal MST from ordinal information alone. For the broad class of doubling metrics, a different algorithm based on \emph{endpoint cut stability} achieves distortion $2^{O(D)}$, where $D$ is the doubling dimension; this is constant for fixed $D$, independently of $n$. Finally, we study a query-based extension allowing comparisons between two edge lengths without revealing their numerical values. For general metrics with $n\geq2$, fewer than $n$ edge-comparison queries suffice to achieve distortion at most $2\log_2^* n-1$, which is near-constant even for very large inputs.

0.25

Almost Optimal Lower Bounds under Almost ETH for Approximating Euclidean k-means

cs.CG cs.CCcs.DSstat.ML ·2026-11-11 ·Karthik C. S., Dor Minzer
theory similarity 0.64
Abstract

In the Euclidean $k$-means problem, we are given a set of $n$ points in $\mathbb{R}^d$, and the goal is to find $k$ representative points in space so as to minimize the sum of the squared distances from each point to its closest representative. Algorithmically, de la Vega, Karpinski, Kenyon, and Rabani [STOC'03] and Kumar, Sabharwal, and Sen [JACM'10] obtain a $(1+\varepsilon)$-approximation for high-dimensional Euclidean $k$-means in time $2^{(k/\varepsilon)^{O(1)}} \cdot dn^{O(1)}$, which was subsequently improved to $2^{\tilde{O}(k/\varepsilon)} + O(ndk)$ by Feldman, Monemizadeh, and Sohler [SoCG'07]. On the lower-bound side, a recent work of Cohen-Addad, Karthik, Saulpic, and Schwiegelshohn [SoCG'26] proposed the Exponential Time for Expanders Hypothesis (XXH), which asserts exponential-time hardness for $(1+\varepsilon)$-approximating Vertex Cover on vertex expanders, and used it to prove nearly matching running time lower bound whenever $k=\tildeω(1/\varepsilon)$. We observe that XXH does not hold in the parameter regime needed for the above result, and show how to repair their reduction under a more standard assumption. More precisely, we show that assuming the Almost Exponential Time Hypothesis for 3-SAT (Almost ETH), for every constant $β\in(0,1)$, Euclidean $k$-means in $\mathbb{R}^d$ has no $(1+\varepsilon)$-approximation algorithm running in time $2^{(k/\varepsilon)^{1-β}}\text{poly}(n,d)$, whenever $k=\tildeω(1/\varepsilon)$. By modifying the construction, we also obtain a similar result for approximating the Euclidean $k$-median problem.

0.12

Metric TSP in Sublinear Time: Breaking Two Approximation

cs.DS ·2026-11-03 ·Amir Azarmehr, Soheil Behnezhad
similarity 0.64
Abstract

The seminal result of Czumaj and Sohler [STOC'04] showed that a 2-approximation of the value of metric TSP can be obtained with sublinear queries. Whether sublinear algorithms can break the 2-approximation barrier has remained a major open problem. See problem 71 on sublinear.info. This paper resolves this open problem. We show there are fixed $\varepsilon, δ> 0$ such that a $(2-δ)$-approximation of metric TSP value can be obtained with $O(n^{2-\varepsilon})$ queries. Here $n$ is the number of vertices, and the input size is $Θ(n^2)$.

0.23

How to Fix a Broken Metric: A Linear Kernel, Tight Bounds, and Tree Metrics

cs.DS ·2026-10-06 ·Sayan Bandyapadhyay, William Lochet, Daniel Lokshtanov, Saket Saurabh, Chinmay Sonar +2 more
hot direction similarity 0.64
Abstract

Given a complete graph whose edge weights represent dissimilarities, Metric Violation Distance asks whether at most $k$ weights can be changed to form a metric. Motivated by metric repair for noisy data, the problem admits a polynomial-time $O(\log n)$-approximation due to Cohen-Addad, Fan, Lee and de Mesmay [SIAM J. Comput., 2025]. In the context of parameterized complexity, Fan, Gilbert, Raichel, Sonthalia and Van Buskirk [SWAT 2020] gave a $k^{O(k)}n^{O(1)}$-time algorithm. Fomin, Golovach and More [IPEC 2026] obtained an $O(k^2)$ kernel and a single-exponential algorithm for the ultrametric case. They asked whether general metrics admit a single-exponential algorithm and a polynomial kernel, and whether the tree-metric analogue is fixed-parameter tractable. We answer all three questions. We give a $2^{O(k)}n^{O(1)}$-time algorithm and prove that, unless ETH fails, no $2^{o(k)}n^{O(1)}$-time algorithm exists, even when all input distances lie in $\{1,2,3\}$. We also give a kernel with at most $6k$ vertices. The kernel supports solution lifting and can precede any approximation algorithm. Combined with the $O(\log n)$-approximation of Cohen-Addad, Fan, Lee and de Mesmay, it yields an $O(\log \mathrm{OPT})$-approximation at no asymptotic cost in running time. Both results extend to an interval generalization in which each edge $e$ has an observed value $M_e$ and an admissible range $[A_e,B_e]$ within which it may be reassigned; the kernel then has $7k$ vertices. Finally, Tree Metric Violation Distance is fixed-parameter tractable and solvable in $k^{O(k)}n^{O(1)}$ time.

0.48