Table of Contents
Fetching ...

Lower Bounds for the Minimum Spanning Tree Cycle Intersection Problem

Manuel Dubinsky, Kun-Mao Chao, César Massri, Gabriel Taubin

TL;DR

The paper investigates the Minimum Spanning Tree Cycle Intersection problem, formalizing the intersection number $\cap(G)$ as the count of non-empty intersections among tree-cycles arising from a spanning tree. It first proves a general lower bound $\frac{1}{2}\left(\frac{\nu^2}{n-1} - \nu\right)$ with $\nu = m - n + 1$, derived via a convex-optimization view over bond contributions, yielding $l_{n,m}$. It then conjectures a tighter bound for graphs with a universal vertex, expressed in terms of $2\nu = q(n-1) + r$ as $\hat{l}_{n,m} = (n-1)\binom{q}{2} + q r \leq \cap(G)$, grounded in the notion of $\nu$-regular graphs. The authors support the conjecture with experiments and discuss implications for the sparsity of the cycle-intersection matrix and potential benefits for fast linear solvers on large graphs. Overall, the work provides foundational lower bounds and a conjectured tight form that could guide both theoretical and practical approaches to MSTCI and related discrete differential forms computations.

Abstract

Minimum spanning trees are important tools in the analysis and design of networks. Many practical applications require their computation, ranging from biology and linguistics to economy and telecommunications. The set of cycles of a network has a vector space structure. Given a spanning tree, the set of non-tree edges defines cycles that determine a basis. The intersection of two such cycles is the number of edges they have in common and the intersection number -- denoted $\cap(G)$ -- is the number of non-empty pairwise intersections of the cycles of the basis. The Minimum Spanning Tree Cycle Intersection problem consists in finding a spanning tree such that the intersection number is minimum. This problem is relevant in order to integrate discrete differential forms. In this paper, we present two lower bounds of the intersection number of an arbitrary connected graph $G=(V,E)$. In the first part, we prove the following statement: $$\frac{1}{2}\left(\frac{ν^2}{n-1} - ν\right) \leq \cap(G),$$ where $n = |V|$ and $ν$ is the \emph{cyclomatic number} of $G$. In the second part, based on some experimental results and a new observation, we conjecture the following improved tight lower bound: $$(n-1) \binom{q}{2} + q \ r\leq \cap(G),$$ where $2 ν= q (n-1) + r$ is the integer division of $2 ν$ and $n-1$. This is the first result in a general context, that is for an arbitrary connected graph.

Lower Bounds for the Minimum Spanning Tree Cycle Intersection Problem

TL;DR

The paper investigates the Minimum Spanning Tree Cycle Intersection problem, formalizing the intersection number as the count of non-empty intersections among tree-cycles arising from a spanning tree. It first proves a general lower bound with , derived via a convex-optimization view over bond contributions, yielding . It then conjectures a tighter bound for graphs with a universal vertex, expressed in terms of as , grounded in the notion of -regular graphs. The authors support the conjecture with experiments and discuss implications for the sparsity of the cycle-intersection matrix and potential benefits for fast linear solvers on large graphs. Overall, the work provides foundational lower bounds and a conjectured tight form that could guide both theoretical and practical approaches to MSTCI and related discrete differential forms computations.

Abstract

Minimum spanning trees are important tools in the analysis and design of networks. Many practical applications require their computation, ranging from biology and linguistics to economy and telecommunications. The set of cycles of a network has a vector space structure. Given a spanning tree, the set of non-tree edges defines cycles that determine a basis. The intersection of two such cycles is the number of edges they have in common and the intersection number -- denoted -- is the number of non-empty pairwise intersections of the cycles of the basis. The Minimum Spanning Tree Cycle Intersection problem consists in finding a spanning tree such that the intersection number is minimum. This problem is relevant in order to integrate discrete differential forms. In this paper, we present two lower bounds of the intersection number of an arbitrary connected graph . In the first part, we prove the following statement: where and is the \emph{cyclomatic number} of . In the second part, based on some experimental results and a new observation, we conjecture the following improved tight lower bound: where is the integer division of and . This is the first result in a general context, that is for an arbitrary connected graph.
Paper Structure (9 sections, 13 theorems, 39 equations, 1 figure)

This paper contains 9 sections, 13 theorems, 39 equations, 1 figure.

Key Result

Lemma 2

Let $e \in E'$ be an edge of $T$ and $b_e$ its corresponding bond. Then, the edges in $b_e - \{e\}$ are cycle-edges that determine tree-cycles that intersect pairwise.

Figures (1)

  • Figure 1: Comparison of mean and standard deviation of $\frac{\hat{l}_{n,m}}{\cap(G)}$ (blue) and $\frac{l_{n,m}}{\cap(G)}$ (red) over a uniformly distributed random sample of size 1000 of $9$-node connected graphs.

Theorems & Definitions (26)

  • Definition 1
  • Lemma 2
  • proof
  • Lemma 3
  • proof
  • Corollary 4
  • proof
  • Definition 5
  • Lemma 6
  • proof
  • ...and 16 more