arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1712.05993math.NAcs.NAmath.OC
收藏

使用矩阵微分方程进行图划分

Graph partitioning using matrix differential equations

Università degli Studi di L’ Aquila(拉奎拉大学);Gran Sasso Science Institute(格兰萨索科学研究所);Universität Tübingen(蒂宾根大学)

Eleonora Andreotti, Dominik Edelmann, Nicola Guglielmi, Christian Lubich

AI总结:

本文研究图划分问题,重点解决最小割问题及约束最小割问题,提出基于矩阵近邻问题的矩阵微分方程约束梯度系统方法。

摘要:

给定一个连通无向加权图,本文关注图划分相关问题。首先寻找最接近的不连通图(最小割问题),此处以欧几里得范数为度量标准。我们感兴趣于具有约束条件的最小割问题,其中约束包括节点数量或成员要求,这导致了NP难的组合优化问题。此外,我们还关注歧义问题,即基于Fiedler谱划分的聚类算法的鲁棒性。上述问题被重新表述为图权重矩阵的矩阵近邻问题。解决这些矩阵近邻问题的关键要素是使用约束梯度系统的矩阵微分方程。

英文摘要:

Given a connected undirected weighted graph, we are concerned with problems related to partitioning the graph. First of all we look for the closest disconnected graph (the minimum cut problem), here with respect to the Euclidean norm. We are interested in the case of constrained minimum cut problems, where constraints include cardinality or membership requirements, which leads to NP-hard combinatorial optimization problems. Furthermore, we are interested in ambiguity issues, that is in the robustness of clustering algorithms that are based on Fiedler spectral partitioning. The above-mentioned problems are restated as matrix nearness problems for the weight matrix of the graph. A key element in the solution of these matrix nearness problems is the use of a constrained gradient system of matrix differential equations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑