SearcharxivSearch

arXiv subjects

Clive Elphick

Publications and source records attributed to Clive Elphick.

At least 19 recordsLinked to original sources

Resolution of a problem of Mohar on non-positive inertia

For a graph $G$ of order $n$, its positive, negative and non-positive inertia is the number of positive, negative and non-positive eigenvalues of its adjacency matrix $A(G)$, respectively. Mohar asked whether every graph with $k$ non-positive eigenvalues has order $O(k^2)$ as $k\to \infty$. Using NEPS, we construct a sequence of non-singular graphs with negative inertia $k$ and order $\Omega(k^{\frac{7}{3}})$ as $k\to \infty$, thus resolving Mohar's problem. Our result also strongly refutes a recent conjecture of Akbari, Elphick, Kumar, Pragada, and Tang involving positive and negative inertia.

math.CO

Proof of a conjectured spectral upper bound on the chromatic number of a graph

Let $G$ be a simple graph on $n$ vertices and $m$ edges with chromatic number $\chi$, and let $\lambda_n$ denote the least adjacency eigenvalue. Solving a conjecture of Fan, Yu and Wang~[Electron. J. Combin., 2012], we prove that when $3\le \chi\le n-1$, the chromatic number satisfies the following upper bound: $$ \chi \le \left(\frac{n}{2}+1+\lambda_n\right) + \sqrt{\left(\frac{n}{2}+1+\lambda_n\right)^{2}-4(\lambda_n+1)\left(\lambda_n+\frac{n}{2}\right)}, $$ with equality if and only if $G \cong \left(K_{\frac{\chi}{2}}\cup\tfrac{n-\chi}{2}K_1\right) \vee \left(K_{\frac{\chi}{2}}\cup\tfrac{n-\chi}{2}K_1\right)$, where both $n$ and $\chi$ are even. This extends the validity of Fan--Yu--Wang's bound from the range $3\le \chi\le \frac{n}{2}$ to the full range $3\le \chi\le n-1$. We also compare this bound with the well-known bound due to Wilf that $\chi \le 1 + \lambda_1$, where $\lambda_1$ denotes the largest eigenvalue. In particular we show that while Wilf's bound is an upper bound for some parameters larger than $\chi$, this bound using $\lambda_n$ is not an upper bound for these parameters. We conclude with a similar conjectured upper bound for $\chi(G)$, which uses $m$ in place of $n$.

math.CO

A new conjecture on the inertia of graphs

Let $G$ be a graph with adjacency matrix $A(G)$. We conjecture that \[2n^+(G) \le n^-(G)(n^-(G) + 1),\] where $n^+(G)$ and $n^-(G)$ denote the number of positive and negative eigenvalues of $A(G)$, respectively. This conjecture generalizes to all graphs the well-known absolute bound for strongly regular graphs. The conjecture also relates to a question posed by Torga\v{s}ev. We prove the conjecture for special graph families, including line graphs and planar graphs, and provide examples where the conjecture is exact. We also conjecture that for any connected graph $G$, its line graph $L(G)$ satisfies $n^+(L(G)) \le n^-(L(G)) + 1$, and obtain partial results.

math.CO

Inertia, Independence and Expanders

Let $G$ be a graph on $n$ vertices, independence number $\alpha(G)$, Lov\'asz theta function $\vartheta(G)$, and Shannon capacity $\Theta(G)$. We define $n_{\ge0}(G)$ to be the minimum number of non-negative eigenvalues taken over all Hermitian weighted adjacency matrices of $G$. It is well known that $\alpha(G) \le \Theta(G) \le \vartheta(G)$ and $\alpha(G) \le n_{\ge0}(G)$. Continuing a long line of work, we investigate the relationships between $ \alpha(G) $, $ \vartheta(G) $, $\Theta(G)$, and $ n_{\ge 0}(G) $. We prove a conjecture of Kwan and Wigderson, showing that for every integer $k$, there exists a graph $G$ with $\alpha(G) \leq 2$ and $n_{\ge 0}(G) \ge k$. In addition, we prove that for every integer $k$, there exists a graph $G$ with $\Theta(G) \leq 3$ and $n_{\ge 0}(G) \ge k$. Both results rely on a new observation: if the complement of $G$ contains a good spectral expander, then $n_{\geq 0}(G)$ must be large. We also show that $\vartheta(G)$ can be exponentially larger than $n_{\ge 0}(G)$, improving a recent result of Ihringer.

math.CO

A Spectral Lower Bound on Chromatic Numbers using $p$-Energy

Let $A_G $ be the adjacency matrix of a simple graph $ G $, and let $ \chi(G) $, $ \chi_f(G) $, $ \chi_q(G) $, $ \xi(G) $ and $ \xi_f(G) $ denote its chromatic number, fractional chromatic number, quantum chromatic number, orthogonal rank and projective rank, respectively. For $ p \geq 0 $, we define the positive and negative $ p $-energies of $ G $ by $$ \mathcal{E}_p^+(G) = \sum_{\lambda_i > 0} \lambda_i^p, \quad \mathcal{E}_p^-(G) = \sum_{\lambda_i < 0} |\lambda_i|^p, $$ where $ \lambda_1 \geq \cdots \geq \lambda_n $ are the eigenvalues of $A_G $. We prove that for all $ p \geq 0 $, $$ \chi(G) \geq \left\{\chi_f(G), \chi_q(G), \xi(G) \right\} \geq \xi_f(G) \geq 1 + \max\left\{ \frac{\mathcal{E}_p^+(G)}{\mathcal{E}_p^-(G)}, \frac{\mathcal{E}_p^-(G)}{\mathcal{E}_p^+(G)} \right\}^{\frac{1}{|p - 1|}}. $$ This result unifies and strengthens a series of existing bounds corresponding to the cases $ p \in \{0, 2, \infty\} $. In particular, the case $ p = 0 $ yields the inertia bound $$ \chi_f(G) \geq \xi_f(G) \geq1 + \max\left\{\frac{n^+}{n^-}, \frac{n^-}{n^+}\right\}, $$ where $ n^+ $ and $ n^- $ denote the number of positive and negative eigenvalues of $ A_G $, respectively. This resolves two conjectures of Elphick and Wocjan. We also demonstrate that for certain graphs, non-integer values of $ p $ provide sharper lower bounds than existing spectral bounds. As an example, we determine $ \chi_q $ for the Tilley graph, which cannot be achieved using existing (unweighted) $p$-energy bounds. Our proof employs a novel synthesis of linear algebra and measure-theoretic tools, which allows us to surpass existing spectral bounds.

math.CO

Symmetry and asymmetry between positive and negative square energies of graphs

The positive and negative square energies of a graph, $s^+(G)$ and $s^-(G)$, are the sums of squares of the positive and negative eigenvalues of the adjacency matrix, respectively. The first results on square energies revealed symmetry between $s^+(G)$ and $s^-(G)$. This paper reviews examples of asymmetry between these parameters, for example using large random graphs and the ratios $s^+/s^-$ and $s^-/s^+$, as well as new examples of symmetry. We answer some questions previously asked about $s^{+}$ and $s^{-}$ and suggest several further avenues of research.

math.CO

Two conjectured strengthenings of Tur\'an's theorem

We investigate two conjectured spectral graph theoretic strengthenings of Tur\'an's theorem. Let $\mu_1 \ge \ldots \ge \mu_n$ denote the eigenvalues of a graph $G$ with $n$ vertices, $m$ edges and clique number $\omega(G)$. The concise version of Tur\'an's theorem is that $n/(n - d)$ is a lower bound for the clique number $\omega(G)$, where $d$ is the average degree. Our first conjecture is that $d$ can be replaced in this bound with $\sqrt{s^+}$, where $s^+$ is the sum of the squares of the positive eigenvalues. We prove this conjecture for triangle-free, weakly perfect and Kneser graphs and for almost all graphs. We have also used various software tools to search for a counter-example. Nikiforov proved a spectral version of Tur\'an's theorem that \[ \mu_1^2 \le \frac{2m(\omega(G) - 1)}{\omega(G)}, \] and Bollob\'as and Nikiforov conjectured that for $G \not = K_n$ \[ \mu_1^2 + \mu_2^2 \le \frac{2m(\omega(G) - 1)}{\omega(G)}. \] For our second conjecture, we propose that for all graphs $(\mu_1^2 + \mu_2^2)$ in this inequality can be replaced by the sum of the squares of the $\omega(G)$ largest eigenvalues, provided they are positive. We prove the conjecture for weakly perfect, Kneser, and classes of strongly regular graphs. We also provide experimental evidence and describe how the bound can be applied. Liu and Ning published a wide-ranging paper entitled ``Unsolved Problems in spectral graph theory'', and these two conjectures were placed second and fourth in their list of such problems.

math.CO

Spectral Lower Bounds for the Quantum Chromatic Number of a Graph -- Part II

Hoffman proved that a graph $G$ with eigenvalues $\mu_1 \ge \ldots \ge \mu_n$ and chromatic number $\chi(G)$ satisfies: \[ \chi \ge 1 + \kappa \] where $\kappa$ is the smallest integer such that \[ \mu_1 + \sum_{i=1}^{\kappa} \mu_{n+1-i} \le 0. \] We strengthen this well known result by proving that $\chi(G)$ can be replaced by the quantum chromatic number, $\chi_q(G)$, where for all graphs $\chi_q(G) \le \chi(G)$ and for some graphs $\chi_q(G)$ is significantly smaller than $\chi(G)$. We also prove a similar result, and investigate implications of these inequalities for the quantum chromatic number of various classes of graphs, which improves many known results. For example, we demonstrate that the Kneser graph $KG_{p,2}$ has $\chi_q = \chi = p - 2$.

math.CO

Spectral upper bound on the quantum k-independence number of a graph

A well known upper bound for the independence number $\alpha(G)$ of a graph $G$, due to Cvetkovi\'{c}, is that \begin{equation*} \alpha(G) \le n^0 + \min\{n^+ , n^-\} \end{equation*} where $(n^+, n^0, n^-)$ is the inertia of $G$. We prove that this bound is also an upper bound for the quantum independence number $\alpha_q$(G), where $\alpha_q(G) \ge \alpha(G)$ and for some graphs $\alpha_q(G) \gg \alpha(G)$. We identify numerous graphs for which $\alpha(G) = \alpha_q(G)$, thus increasing the number of graphs for which $\alpha_q$ is known. We also demonstrate that there are graphs for which the above bound is not exact with any Hermitian weight matrix, for $\alpha(G)$ and $\alpha_q(G)$. Finally, we show this result in the more general context of spectral bounds for the quantum $k$-independence number, where the $k$-independence number is the maximum size of a set of vertices at pairwise distance greater than $k$.

math.CO

More Tales of Hoffman: bounds for the vector chromatic number of a graph

Let $\chi(G)$ denote the chromatic number of a graph and $\chi_v(G)$ denote the vector chromatic number. For all graphs $\chi_v(G) \le \chi(G)$ and for some graphs $\chi_v(G) \ll \chi(G)$. Galtman proved that Hoffman's well-known lower bound for $\chi(G)$ is in fact a lower bound for $\chi_v(G)$. We prove that two more spectral lower bounds for $\chi(G)$ are also lower bounds for $\chi_v(G)$. We then use one of these bounds to derive a new characterization of $\chi_v(G)$.

math.CO

An inertial upper bound for the quantum independence number of a graph

A well known upper bound for the independence number $\alpha(G)$ of a graph $G$, is that \[ \alpha(G) \le n^0 + \min\{n^+ , n^-\}, \] where $(n^+, n^0, n^-)$ is the inertia of $G$. We prove that this bound is also an upper bound for the quantum independence number $\alpha_q$(G), where $\alpha_q(G) \ge \alpha(G)$. We identify numerous graphs for which $\alpha(G) = \alpha_q(G)$ and demonstrate that there are graphs for which the above bound is not exact with any Hermitian weight matrix, for $\alpha(G)$ and $\alpha_q(G)$. This result complements results by the authors that many spectral lower bounds for the chromatic number are also lower bounds for the quantum chromatic number.

math.CO

Spectral lower bounds for the orthogonal and projective ranks of a graph

The orthogonal rank of a graph $G=(V,E)$ is the smallest dimension $\xi$ such that there exist non-zero column vectors $x_v\in\mathbb{C}^\xi$ for $v\in V$ satisfying the orthogonality condition $x_v^\dagger x_w=0$ for all $vw\in E$. We prove that many spectral lower bounds for the chromatic number, $\chi$, are also lower bounds for $\xi$. This result complements a previous result by the authors, in which they showed that spectral lower bounds for $\chi$ are also lower bounds for the quantum chromatic number $\chi_q$. It is known that the quantum chromatic number and the orthogonal rank are incomparable. We conclude by proving an inertial lower bound for the projective rank $\xi_f$, and conjecture that a stronger inertial lower bound for $\xi$ is also a lower bound for $\xi_f$.

math.CO

Spectral lower bounds for the quantum chromatic number of a graph

The quantum chromatic number, $\chi_q(G)$, of a graph $G$ was originally defined as the minimal number of colors necessary in a quantum protocol in which two provers that cannot communicate with each other but share an entangled state can convince an interrogator with certainty that they have a coloring of the graph. We use an equivalent purely combinatorial definition of $\chi_q(G)$ to prove that many spectral lower bounds for the chromatic number, $\chi(G)$, are also lower bounds for $\chi_q(G)$. This is achieved using techniques from linear algebra called pinching and twirling. We illustrate our results with some examples.

math.CO

Conjectured lower bound for the clique number of a graph

It is well known that $n/(n - \mu)$, where $\mu$ is the spectral radius of a graph with $n$ vertices, is a lower bound for the clique number. We conjecture that $\mu$ can be replaced in this bound with $\sqrt{s^+}$, where $s^+$ is the sum of the squares of the positive eigenvalues. We prove this conjecture for various classes of graphs, including triangle-free graphs, and for almost all graphs.

math.CO

Conjectured bound for the distribution of eigenvalues of a graph

Let $(n^+, n^0, n^-)$ denote the inertia of a graph $G$ with $n$ vertices. Nordhaus-Gaddum bounds are known for inertia, except for an upper bound for $n^-$. We conjecture that for any graph \[ n^-(G) + n^-(\bar{G}) \le 1.5(n - 1), \] and prove this bound for various classes of graphs and for almost all graphs. We consider the relationship between this bound and the number of eigenvalues that lie within the interval $-1$ to $0$, which we denote $n_{(-1,0)}(G)$. We conjecture that for any graph \[ n_{(-1,0)}(G) \le 0.5(n - 1). \] and prove this bound for almost all graphs. We also investigate extremal graphs for both bounds and show that both bounds are equivalent for regular graphs.

math.CO

Nordhaus-Gaddum and other bounds for the sum of squares of the positive eigenvalues of a graph

Terpai [22] proved the Nordhaus-Gaddum bound that $\mu(G) + \mu(\overline{G}) \le 4n/3 - 1$, where $\mu(G)$ is the spectral radius of a graph $G$ with $n$ vertices. Let $s^+$ denote the sum of the squares of the positive eigenvalues of $G$. We prove that $\sqrt{s^{+}(G)} + \sqrt{s^+(\overline{G})} < \sqrt{2}n$ and conjecture that $\sqrt{s^{+}(G)} + \sqrt{s^+(\overline{G})} \le 4n/3 - 1.$ We have used AutoGraphiX and Wolfram Mathematica to search for a counter-example. We also consider Nordhaus-Gaddum bounds for $s^+$ and bounds for the Randi\'c index.

math.CO

An inertial lower bound for the chromatic number of a graph

Let $\chi(G$) and $\chi_f(G)$ denote the chromatic and fractional chromatic numbers of a graph $G$, and let $(n^+ , n^0 , n^-)$ denote the inertia of $G$. We prove that: \[ 1 + \max\left(\frac{n^+}{n^-} , \frac{n^-}{n^+}\right) \le \chi(G) \mbox{ and conjecture that } 1 + \max\left(\frac{n^+}{n^-} , \frac{n^-}{n^+}\right) \le \chi_f(G) \] We investigate extremal graphs for these bounds and demonstrate that this inertial bound is not a lower bound for the vector chromatic number. We conclude with a discussion of asymmetry between $n^+$ and $n^-$, including some Nordhaus-Gaddum bounds for inertia.

math.CO

Upper bounds for the achromatic and coloring numbers of a graph

Dvo\v{r}\'ak \emph{et al.} introduced a variant of the Randi\'c index of a graph $G$, denoted by $R'(G)$, where $R'(G)=\sum_{uv\in E(G)}\frac 1 {\max\{d(u), d(v)\}}$, and $d(u)$ denotes the degree of a vertex $u$ in $G$. The coloring number $col(G)$ of a graph $G$ is the smallest number $k$ for which there exists a linear ordering of the vertices of $G$ such that each vertex is preceded by fewer than $k$ of its neighbors. It is well-known that $\chi(G)\leq col(G)$ for any graph $G$, where $\chi(G)$ denotes the chromatic number of $G$. In this note, we show that for any graph $G$ without isolated vertices, $col(G)\leq 2R'(G)$, with equality if and only if $G$ is obtained from identifying the center of a star with a vertex of a complete graph. This extends some known results. In addition, we present some new spectral bounds for the coloring and achromatic numbers of a graph.

math.CO