English

Query Efficient Weighted Stochastic Matching

Data Structures and Algorithms 2023-11-16 v1

Abstract

In this paper, we study the weighted stochastic matching problem. Let G=(V,E)G=(V, E) be a given edge-weighted graph and let its realization G\mathcal{G} be a random subgraph of GG that includes each edge eEe\in E independently with a known probability pep_e. The goal in this problem is to pick a sparse subgraph QQ of GG without prior knowledge of GG's realization, such that the maximum weight matching among the realized edges of QQ (i.e. the subgraph QGQ\cap \mathcal{G}) in expectation approximates the maximum weight matching of the entire realization G\mathcal{G}. Attaining any constant approximation ratio for this problem requires selecting a subgraph of max-degree Ω(1/p)\Omega(1/p) where p=mineEpep=\min_{e\in E} p_e. On the positive side, there exists a (1ϵ)(1-\epsilon)-approximation algorithm by Behnezhad and Derakhshan, albeit at the cost of max-degree having exponential dependence on 1/p1/p. Within the poly(1/p)\text{poly}(1/p) regime, however, the best-known algorithm achieves a 0.5360.536 approximation ratio due to Dughmi, Kalayci, and Patel improving over the 0.5010.501 approximation algorithm by Behnezhad, Farhadi, Hajiaghayi, and Reyhani. In this work, we present a 0.68 approximation algorithm with O(1/p)O(1/p) queries per vertex, which is asymptotically tight. This is even an improvement over the best-known approximation ratio of 2/32/3 for unweighted graphs within the poly(1/p)\text{poly}(1/p) regime due to Assadi and Bernstein. The 2/32/3 approximation ratio is proven tight in the presence of a few correlated edges in G\mathcal{G}, indicating that surpassing the 2/32/3 barrier should rely on the independent realization of edges. Our analysis involves reducing the problem to designing a randomized matching algorithm on a given stochastic graph with some variance-bounding properties.

Keywords

Cite

@article{arxiv.2311.08513,
  title  = {Query Efficient Weighted Stochastic Matching},
  author = {Mahsa Derakhshan and Mohammad Saneian},
  journal= {arXiv preprint arXiv:2311.08513},
  year   = {2023}
}
R2 v1 2026-06-28T13:21:19.464Z