arXiv subjects

L. Sunil Chandran

Publications and source records attributed to L. Sunil Chandran.

88 records · Page 5Linked to original sources

arXiv2008

$d$-Regular graphs of acyclic chromatic index at least $d+2$

An $acyclic$ edge coloring of a graph is a proper edge coloring such that there are no bichromatic cycles. The \emph{acyclic chromatic index} of a graph is the minimum number k such that there is an acyclic edge coloring using k colors and is denoted by $a'(G)$. It was conjectured by Alon, Sudakov and Zaks (and earlier by Fiamcik) that $a'(G)\le Δ+2$, where $Δ=Δ(G)$ denotes the maximum degree of the graph. Alon et.al also raised the question whether the complete graphs of even order are the only regular graphs which require $Δ+2$ colors to be acyclically edge colored. In this paper, using a simple counting argument we observe not only that this is not true, but infact all d-regular graphs with $2n$ vertices and $d > n$, requires at least $d+2$ colors. We also show that $a'(K_{n,n}) \ge n+2$, when $n$ is odd using a more non-trivial argument(Here $K_{n,n}$ denotes the complete bipartite graph with $n$ vertices on each side). This lower bound for $K_{n,n}$ can be shown to be tight for some families of complete bipartite graphs and for small values of $n$. We also infer that for every $d,n$ such that $d \ge 5$, $n \ge 2d + 3$ and $dn$ even, there exist $d$-regular graphs which require at least $d+2$-colors to be acyclically edge colored.

math.CO↗

arXiv2008

On the cubicity of AT-free graphs and circular-arc graphs

A unit cube in $k$ dimensions ($k$-cube) is defined as the the Cartesian product $R_1\times R_2\times...\times R_k$ where $R_i$(for $1\leq i\leq k$) is a closed interval of the form $[a_i,a_i+1]$ on the real line. A graph $G$ on $n$ nodes is said to be representable as the intersection of $k$-cubes (cube representation in $k$ dimensions) if each vertex of $G$ can be mapped to a $k$-cube such that two vertices are adjacent in $G$ if and only if their corresponding $k$-cubes have a non-empty intersection. The \emph{cubicity} of $G$ denoted as $\cubi(G)$ is the minimum $k$ for which $G$ can be represented as the intersection of $k$-cubes. We give an $O(bw\cdot n)$ algorithm to compute the cube representation of a general graph $G$ in $bw+1$ dimensions given a bandwidth ordering of the vertices of $G$, where $bw$ is the \emph{bandwidth} of $G$. As a consequence, we get $O(Δ)$ upper bounds on the cubicity of many well-known graph classes such as AT-free graphs, circular-arc graphs and co-comparability graphs which have $O(Δ)$ bandwidth. Thus we have: 1) $\cubi(G)\leq 3Δ-1$, if $G$ is an AT-free graph. 2) $\cubi(G)\leq 2Δ+1$, if $G$ is a circular-arc graph. 3) $\cubi(G)\leq 2Δ$, if $G$ is a co-comparability graph. Also for these graph classes, there are constant factor approximation algorithms for bandwidth computation that generate orderings of vertices with $O(Δ)$ width. We can thus generate the cube representation of such graphs in $O(Δ)$ dimensions in polynomial time.

cs.DM↗

arXiv2008

Representing graphs as the intersection of axis-parallel cubes

A unit cube in $k$ dimensional space (or \emph{$k$-cube} in short) is defined as the Cartesian product $R_1\times R_2\times...\times R_k$ where $R_i$(for $1\leq i\leq k$) is a closed interval of the form $[a_i,a_i+1]$ on the real line. A $k$-cube representation of a graph $G$ is a mapping of the vertices of $G$ to $k$-cubes such that two vertices in $G$ are adjacent if and only if their corresponding $k$-cubes have a non-empty intersection. The \emph{cubicity} of $G$, denoted as $\cubi(G)$, is the minimum $k$ such that $G$ has a $k$-cube representation. Roberts \cite{Roberts} showed that for any graph $G$ on $n$ vertices, $\cubi(G)\leq 2n/3$. Many NP-complete graph problems have polynomial time deterministic algorithms or have good approximation ratios in graphs of low cubicity. In most of these algorithms, computing a low dimensional cube representation of the given graph is usually the first step. We present an efficient algorithm to compute the $k$-cube representation of $G$ with maximum degree $Δ$ in $O(Δ\ln b)$ dimensions where $b$ is the bandwidth of $G$. Bandwidth of $G$ is at most $n$ and can be much lower. The algorithm takes as input a bandwidth ordering of the vertices in $G$. Though computing the bandwidth ordering of vertices for a graph is NP-hard, there are heuristics that perform very well in practice. Even theoretically, there is an $O(\log^4 n)$ approximation algorithm for computing the bandwidth ordering of a graph using which our algorithm can produce a $k$-cube representation of any given graph in $k=O(Δ(\ln b + \ln\ln n))$ dimensions. Both the bounds on cubicity are shown to be tight upto a factor of $O(\log\log n)$.

cs.DM↗

arXiv2008

Acyclic Edge Coloring of Graphs with Maximum Degree 4

math.CO↗

arXiv2008

Bounds On Isoperimetric Values of Trees

Let G = (V,E) be a finite, simple and undirected graph. For $S \subseteq V$, let $δ(S,G) = \{(u,v) \in E : u \in S \mbox {and} v \in V-S \}$ be the edge boundary of $S$. Given an integer $i$, $1 \leq i \leq | V |$, let the edge isoperimetric value of $G$ at $i$ be defined as $b_e(i,G) = \min_{S \subseteq V; |S| = i} |δ(S,G)|$. The edge isoperimetric peak of $G$ is defined as $b_e(G)=\max_{1 \leq j \leq | V |} b_e(j,G)$. Let $b_v(G)$ denote the vertex isoperimetric peak defined in a corresponding way. The problem of determining a lower bound for the vertex isoperimetric peak in complete $t$-ary trees was recently considered in \cite{OatYam}. In this paper we provide bounds which improve those in \cite{OatYam}. We show that for a complete binary tree of depth $d$ (denoted as $T_d^2$), $c_1d \leq b_e(T_d^2) \leq d$ and $c_2d \leq b_v(T_d^2) \leq d$ where $c_1$, $c_2$ are constants. For a complete $t$-ary tree of depth $d$ (denoted as $T_d^t$) and $d \geq c\log{t}$ where $c$ is a constant, we show that $c_1\sqrt{t}d \leq b_e(T_d^t) \leq td$ and $c_2\frac{d}{\sqrt{t}} \leq b_v(T_d^t) \leq d$ where $c_1$, $c_2$ are constants. Our results are generalized to arbitrary(rooted) trees.

math.CO↗

arXiv2007

Cubicity, Boxicity and Vertex Cover

A $k$-dimensional box is the cartesian product $R_1 \times R_2 \times ... \times R_k$ where each $R_i$ is a closed interval on the real line. The {\it boxicity} of a graph $G$, denoted as $box(G)$, is the minimum integer $k$ such that $G$ is the intersection graph of a collection of $k$-dimensional boxes. A unit cube in $k$-dimensional space or a $k$-cube is defined as the cartesian product $R_1 \times R_2 \times ... \times R_k$ where each $R_i$ is a closed interval on the real line of the form $[a_i, a_{i}+1]$. The {\it cubicity} of $G$, denoted as $cub(G)$, is the minimum $k$ such that $G$ is the intersection graph of a collection of $k$-cubes. In this paper we show that $cub(G) \leq t + \left \lceil \log (n - t)\right\rceil - 1$ and $box(G) \leq \left \lfloor\frac{t}{2}\right\rfloor + 1$, where $t$ is the cardinality of the minimum vertex cover of $G$ and $n$ is the number of vertices of $G$. We also show the tightness of these upper bounds. F. S. Roberts in his pioneering paper on boxicity and cubicity had shown that for a graph $G$, $box(G) \leq \left \lfloor\frac{n}{2} \right \rfloor$, where $n$ is the number of vertices of $G$, and this bound is tight. We show that if $G$ is a bipartite graph then $box(G) \leq \left \lceil\frac{n}{4} \right\rceil$ and this bound is tight. We point out that there exist graphs of very high boxicity but with very low chromatic number. For example there exist bipartite (i.e., 2 colorable) graphs with boxicity equal to $\frac{n}{4}$. Interestingly, if boxicity is very close to $\frac{n}{2}$, then chromatic number also has to be very high. In particular, we show that if $box(G) = \frac{n}{2} - s$, $s \geq 0$, then $χ(G) \geq \frac{n}{2s+2}$, where $χ(G)$ is the chromatic number of $G$.

math.CO↗

arXiv2007

Hadwiger Number and the Cartesian Product Of Graphs

The Hadwiger number mr(G) of a graph G is the largest integer n for which the complete graph K_n on n vertices is a minor of G. Hadwiger conjectured that for every graph G, mr(G) >= chi(G), where chi(G) is the chromatic number of G. In this paper, we study the Hadwiger number of the Cartesian product G [] H of graphs. As the main result of this paper, we prove that mr(G_1 [] G_2) >= h\sqrt{l}(1 - o(1)) for any two graphs G_1 and G_2 with mr(G_1) = h and mr(G_2) = l. We show that the above lower bound is asymptotically best possible. This asymptotically settles a question of Z. Miller (1978). As consequences of our main result, we show the following: 1. Let G be a connected graph. Let the (unique) prime factorization of G be given by G_1 [] G_2 [] ... [] G_k. Then G satisfies Hadwiger's conjecture if k >= 2.log(log(chi(G))) + c', where c' is a constant. This improves the 2.log(chi(G))+3 bound of Chandran and Sivadasan. 2. Let G_1 and G_2 be two graphs such that chi(G_1) >= chi(G_2) >= c.log^{1.5}(chi(G_1)), where c is a constant. Then G_1 [] G_2 satisfies Hadwiger's conjecture. 3. Hadwiger's conjecture is true for G^d (Cartesian product of G taken d times) for every graph G and every d >= 2. This settles a question by Chandran and Sivadasan (They had shown that the Hadiwger's conjecture is true for G^d if d >= 3.)

math.CO↗

arXiv2007

Boxicity of Halin Graphs

A k-dimensional box is the Cartesian product R_1 x R_2 x ... x R_k where each R_i is a closed interval on the real line. The boxicity of a graph G, denoted as box(G) is the minimum integer k such that G is the intersection graph of a collection of k-dimensional boxes. Halin graphs are the graphs formed by taking a tree with no degree 2 vertex and then connecting its leaves to form a cycle in such a way that the graph has a planar embedding. We prove that if G is a Halin graph that is not isomorphic to K_4, then box(G)=2. In fact, we prove the stronger result that if G is a planar graph formed by connecting the leaves of any tree in a simple cycle, then box(G)=2 unless G is isomorphic to K_4 (in which case its boxicity is 1).

math.CO↗

arXiv2007

Geometric representation of graphs in low dimension

We give an efficient randomized algorithm to construct a box representation of any graph G on n vertices in $1.5 (Δ+ 2) \ln n$ dimensions, where $Δ$ is the maximum degree of G. We also show that $\boxi(G) \le (Δ+ 2) \ln n$ for any graph G. Our bound is tight up to a factor of $\ln n$. We also show that our randomized algorithm can be derandomized to get a polynomial time deterministic algorithm. Though our general upper bound is in terms of maximum degree $Δ$, we show that for almost all graphs on n vertices, its boxicity is upper bound by $c\cdot(d_{av} + 1) \ln n$ where d_{av} is the average degree and c is a small constant. Also, we show that for any graph G, $\boxi(G) \le \sqrt{8 n d_{av} \ln n}$, which is tight up to a factor of $b \sqrt{\ln n}$ for a constant b.

cs.DM↗

arXiv2006

Boxicity and Maximum degree

An axis-parallel $d$--dimensional box is a Cartesian product $R_1 \times R_2 \times ... \times R_d$ where $R_i$ (for $1 \le i \le d$) is a closed interval of the form $[a_i, b_i]$ on the real line. For a graph $G$, its \emph{boxicity} $\boxi(G)$ is the minimum dimension $d$, such that $G$ is representable as the intersection graph of (axis--parallel) boxes in $d$--dimensional space. The concept of boxicity finds applications in various areas such as ecology, operation research etc. We show that for any graph $G$ with maximum degree $Δ$, $\boxi(G) \le 2 Δ^2 + 2$. That the bound does not depend on the number of vertices is a bit surprising considering the fact that there are highly connected bounded degree graphs such as expander graphs. Our proof is very short and constructive. We conjecture that $\boxi(G)$ is $O(Δ)$.

math.CO↗

arXiv2006

A Combinatorial Family of Near Regular LDPC Codes

An elementary combinatorial Tanner graph construction for a family of near-regular low density parity check codes achieving high girth is presented. The construction allows flexibility in the choice of design parameters like rate, average degree, girth and block length of the code and yields an asymptotic family. The complexity of constructing codes in the family grows only quadratically with the block length.

cs.IT↗

arXiv2006

Hadwiger's Conjecture for Proper Circular Arc Graphs

Circular arc graphs are graphs whose vertices can be represented as arcs on a circle such that any two vertices are adjacent if and only if their corresponding arcs intersect. Proper circular arc graphs are graphs which have a circular arc representation where no arc is completely contained in any other arc. Hadwiger's conjecture states that if a graph $G$ has chromatic number $k$, then a complete graph of $k$ vertices is a minor of $G$. We prove Hadwiger's conjecture for proper circular arc graphs.

math.CO↗

arXiv2006

An upper bound for Cubicity in terms of Boxicity

An axis-parallel b-dimensional box is a Cartesian product $R_1 \times R_2 \times ... \times R_b$ where each $R_i$ (for $1 \leq i \leq b$) is a closed interval of the form $[a_i,b_i]$ on the real line. The boxicity of any graph $G$, box(G) is the minimum positive integer b such that G can be represented as the intersection graph of axis parallel b-dimensional boxes. A b-dimensional cube is a Cartesian product $R_1 \times R_2\times ... \times R_b$, where each $R_i$ (for $1 \leq i \leq b$) is a closed interval of the form [$a_i$,$a_i$+1] on the real line. When the boxes are restricted to be axis-parallel cubes in b-dimension, the minimum dimension b required to represent the graph is called the cubicity of the graph (denoted by cub(G)). In this paper we prove that $cub(G)\leq \lceil \log n \rceil \boxi(G)$} where n is the number of vertices in the graph. This upper bound is tight.

math.CO↗

arXiv2006

On the Boxicity and Cubicity of Hypercubes

For a graph $G$, its \emph{cubicity} $cub(G)$ is the minimum dimension $k$ such that $G$ is representable as the intersection graph of (axis--parallel) cubes in $k$--dimensional space. Chandran, Mannino and Oriolo showed that for a $d$--dimensional hypercube $H_d$, $\frac{d-1}{\log d} \le cub(H_d) \le 2d$. In this paper, we show that $cub(H_d) = Θ(\frac{d}{\log d})$.The parameter \emph{boxicity} generalizes cubicity: the boxicity $box(G)$ of a graph $G$ is defined as the minimum dimension $k$ such that $G$ is representable as the intersection graph of axis parallel boxes in $k$ dimensional space. Since $box(G) \le cub(G)$ for any graph $G$, our result implies that $box(H_d) = O(\frac{d}{\log d})$. The problem of determining a non-trivial lower bound for $box(H_d)$ is left open.

math.CO↗

arXiv2005

Boxicity of Series Parallel Graphs

The three well-known graph classes, planar graphs (P), series-parallel graphs(SP) and outer planar graphs(OP) satisfy the following proper inclusion relation: OP C SP C P. It is known that box(G) <= 3 if G belongs to P and box(G) <= 2 if G belongs to OP. Thus it is interesting to decide whether the maximum possible value of the boxicity of series-parallel graphs is 2 or 3. In this paper we construct a series-parallel graph with boxicity 3, thus resolving this question. Recently Chandran and Sivadasan showed that for any G, box(G) <= treewidth(G)+2. They conjecture that for any k, there exists a k-tree with boxicity k+1. (This would show that their upper bound is tight but for an additive factor of 1, since the treewidth of any k-tree equals k.) The series-parallel graph we construct in this paper is a 2-tree with boxicity 3 and is thus a first step towards proving their conjecture.

math.CO↗

arXiv2005

Boxicity and Treewidth

In this paper, we relate the seemingly unrelated concepts of treewidth and boxicity. Our main result is that, for any graph G, boxicity(G) <= treewidth(G) + 2. We also show that this upper bound is (almost) tight. Our result leads to various interesting consequences, like bounding the boxicity of many well known graph classes, such as chordal graphs, circular arc graphs, AT-free graphs, co--comparability graphs etc. All our bounds are shown to be tight up to small constant factors. An algorithmic consequence of our result is a linear time algorithm to construct a box representation for graphs of bounded treewidth in a space of constant dimension. We also show many structural results as a consequence. In particular, we show that, if the boxicity of a graph is b >= 3, then there exists a simple cycle of length at least b-3 as well as an induced cycle of length at least floor of (log(b-2) to the base Delta) + 2, where Delta is its maximum degree. We also relate boxicity with the cardinality of minimum vertex cover, minimum feedback vertex cover etc. Another structural consequence is that, for any fixed planar graph H, there is a constant c(H) such that, if boxicity(G) >= c(H) then H is a minor of G.

math.CO↗