Table of Contents
Fetching ...

Strengthening Wilf's lower bound on clique number

Hareshkumar Jadav, Sreekara Madyastha, Rahul Raut, Ranveer Singh

TL;DR

This work targets strengthening Wilf's lower bound on the clique number by validating the Elphick–Wocjan conjecture $\frac{n}{n-\sqrt{s^{+}}}\leq \omega$ for several graph classes. The authors analyze eigenvalue structures of conference graphs, strongly regular graphs with $\lambda=\mu$, line graphs $L(K_n)$, Cartesian products of SRGs, and Ramanujan graphs, leveraging $s^{+}=\sum_{\lambda_i>0}\lambda_i^{2}$ to bound $\omega$. Key results include proving the conjecture for conference graphs and $srg(n,d,\lambda,\mu)$ with $\lambda=\mu$ and $n\ge 2d$, establishing the conjecture for $L(K_n)$, showing closure under Cartesian product for SRGs when $n>7$, and confirming the conjecture for Ramanujan graphs with $n\ge 11d$. These findings advance spectral bounds on clique numbers for several structurally important graph families and guide future work toward the general case.

Abstract

Given an integer $k$, deciding whether a graph has a clique of size $k$ is an NP-complete problem. Wilf's inequality provides a spectral bound for the clique number of simple graphs. Wilf's inequality is stated as follows: $\frac{n}{n - λ_{1}} \leq ω$, where $λ_1$ is the largest eigenvalue of the adjacency matrix $A(G)$, $n$ is the number of vertices in $G$, and $ω$ is the clique number of $G$. Strengthening this bound, Elphick and Wocjan proposed a conjecture in 2018, which is stated as follows: $\frac{n}{n - \sqrt{s^{+}}} \leq ω$, where $s^+ = \sum_{λ_{i} > 0} λ_{i}^2$ and $λ_i$ are the eigenvalues of $A(G)$. In this paper, we have settled this conjecture for some classes of graphs, such as conference graphs, strongly regular graphs with $λ= μ$ (i.e., $srg(n, d, μ, μ)$) and $n\geq 2d$, the line graph of $K_{n}$, the Cartesian product of strongly regular graphs, and Ramanujan graph with $n\geq 11d$.

Strengthening Wilf's lower bound on clique number

TL;DR

This work targets strengthening Wilf's lower bound on the clique number by validating the Elphick–Wocjan conjecture for several graph classes. The authors analyze eigenvalue structures of conference graphs, strongly regular graphs with , line graphs , Cartesian products of SRGs, and Ramanujan graphs, leveraging to bound . Key results include proving the conjecture for conference graphs and with and , establishing the conjecture for , showing closure under Cartesian product for SRGs when , and confirming the conjecture for Ramanujan graphs with . These findings advance spectral bounds on clique numbers for several structurally important graph families and guide future work toward the general case.

Abstract

Given an integer , deciding whether a graph has a clique of size is an NP-complete problem. Wilf's inequality provides a spectral bound for the clique number of simple graphs. Wilf's inequality is stated as follows: , where is the largest eigenvalue of the adjacency matrix , is the number of vertices in , and is the clique number of . Strengthening this bound, Elphick and Wocjan proposed a conjecture in 2018, which is stated as follows: , where and are the eigenvalues of . In this paper, we have settled this conjecture for some classes of graphs, such as conference graphs, strongly regular graphs with (i.e., ) and , the line graph of , the Cartesian product of strongly regular graphs, and Ramanujan graph with .

Paper Structure

This paper contains 6 sections, 6 theorems, 50 equations.

Key Result

Theorem 2.1

If a graph $G$ is conference graph, then

Theorems & Definitions (13)

  • Conjecture 1.1
  • Theorem 2.1
  • proof
  • Theorem 2.2
  • proof
  • Theorem 3.1
  • proof
  • Lemma 4.1
  • proof
  • Theorem 4.1
  • ...and 3 more