arXiv Science⌕ Search
Home/Archive/arXiv

arXiv · 2505.18508

Performance report of heuristic algorithm that cracked the largest Gset Ising problems (G81 cut=14060)

Kenneth M. Zick
View original source ↗Read source document ↗Export metadata ↓

Abstract

For the past 25 years, the Gset benchmark problems have challenged all manner of Ising and Max-Cut solvers. The largest of these problems have remained unsolved by any heuristic algorithm. In this report we provide data showing dramatically better speed and accuracy on these large sparse problems. Our newly discovered heuristic algorithm called Cosm reaches high (99.9% of best) solution quality orders of magnitude faster than the previous best heuristic solver results. Additionally, when afforded enough steps Cosm attains higher cuts than ever previously reported, specifically on instances G72 (cut=7008), G77 (cut=9940), and the 20,000-variable G81 (cut=14060). This report includes solution bitstrings so that the cuts can be independently validated. Remarkably, the new best solutions appear to be optimal. We believe the results are an early hint of disruptive opportunities for unconventional, hardware-centric approaches to algorithm discovery.

Explore related subjects

cs.DScond-mat.dis-nncs.ET

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Kenneth M. Zick. 2025-05-24. Performance report of heuristic algorithm that cracked the largest Gset Ising problems (G81 cut=14060). https://arxiv.org/abs/2505.18508

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

The Marked Edge Walk: A Novel MCMC Algorithm for Sampling of Graph Partitions

Novel Markov Chain Monte Carlo (MCMC) methods have enabled the generation of large ensembles of redistricting plans modeled as a graph partitioning problem. However, existing algorithms such as Reversible Recombination (RevReCom) and Metropolized Forest Recombination (MFR) have strong preferences for distributions related to the spanning tree measure. In this paper we introduce the Marked Edge Walk (MEW), a novel Markov chain proposal for sampling from the space of graph partitions. The walk operates on the space of spanning trees with marked edges, allowing for calculable transition probabilities for use in the Metropolis-Hastings algorithm. Empirical results on real-world dual graphs show convergence under a broad class of target distributions less constrained by spanning tree counts, including policy-based distributions, such as competitiveness on New Hampshire that are independent of spanning trees, and compactness and partisan symmetry distributions on New Hampshire and Texas that, while related to spanning trees, can now be properly targeted with a smaller degree of spanning tree bias, which represents an advancement in flexible ensemble generation.

cs.DS↗
arXiv2026

The Price of Privacy For Approximating Max-CSP

We study approximation algorithms for Maximum Constraint Satisfaction Problems (Max-CSPs) under differential privacy (DP) where the constraints are considered sensitive data. Information-theoretically, we aim to classify the best approximation ratios possible for a given privacy budget $\varepsilon$. In the high-privacy regime ($\varepsilon \ll 1$), we show that any $\varepsilon$-DP algorithm cannot beat a random assignment by more than $O(\varepsilon)$ in the approximation ratio. We devise a polynomial-time algorithm which matches this barrier under the assumptions that the instances are bounded-degree and triangle-free. Finally, we show that one or both of these assumptions can be removed for specific CSPs--such as Max-Cut or Max $k$-XOR--albeit at the cost of computational efficiency.

cs.DS↗
arXiv2026

Distributed Santa Claus via Global Rounding

In this paper, we initiate the study of a new class of problems in the CONGEST model: Mixed packing and covering linear programs (LP). Previously, the design of optimization algorithms in the distributed setting was heavily focused on solving linear programs with either packing or covering constraints. We are the first to explore the class of linear programs with both packing and covering constraints by providing a general-purpose CONGEST solver for such LPs and by studying the sequentially well-studied Santa Claus problem as a central representative. This NP-hard problem can be modeled as a bipartite graph of children and gifts where an edge indicates that a child desires a gift. The goal is to assign the gifts to the children such that the least happy child is as happy as possible. Even though this is a well-studied problem in the sequential setting, we provide the first results in the distributed setting. In particular, we show that the complexity of computing an $\mathcal{O}(\log n/\log \log n)$-approximation is $\widehat Θ(\sqrt n+D)$ rounds, where our $\widetildeΩ(\sqrt n+D)$-round lower bound even holds for any approximation.

cs.DS↗