Search

‡ Abstract. In this short paper we study the game of cops and robbers, which is played on the vertices of some fixed graph G. Cops and a robber are allowed to move along the edges of G, and the goal of cops is to capture the robber. The cop number cðGÞ of G is the minimum number of cops required to win the game. Meyniel conjectured a long time ago that Oð ffiffiffi p Þ cops are enough for any connected G on n vertices. Improving several previous results, we prove that the cop number of an n-vertex graph is at most n2 −ð1þoð1ÞÞ ffiffiffiffiffiffiffiffi log n p .A similar result independently and slightly before us was also obtained by Lu and Peng.

In this paper we introduce a natural model for the realization space of a polytope up to projective equivalence which we call the slack realization space of the polytope. The model arises from the positive part of an algebraic variety determined by the slack ideal of the polytope. This is a saturated determinantal ideal that encodes the combinatorics of the polytope. We also derive a new model of the realization space of a polytope from the positive part of the variety of a related ideal. The slack ideal offers an effective computational framework for several classical questions about polytopes such as rational realizability, non-prescribability of faces, and realizability of combinatorial polytopes., Comment: The original arXiv submission of this paper has now been split into two parts; this paper describing the slack realization space of a polytope and applications of the slack ideal, and a second paper focussing on toric slack ideals and projective uniqueness

Kavut, Selçuk (Balikesir Author), Very recently, a class of cryptographically significant Boolean functions were constructed by Tang and Maitra [IEEE Trans. Inform. Theory, 64 ( 2018), pp. 393-402] by modifying the PSap class of bent functions. The basic ideas used in Tang-Maitra construction were derived from a modification of a subclass of bent functions which is defined over the finite field, and a concern was raised in the same paper whether the implementation of such functions will be as efficient as that of Maiorana-McFarland type bent functions. In this paper, we look at the concrete realization of such functions over a vector space and answer the question positively. The first part of this paper investigates how the finite field implementation of the functions can be viewed as simple truth tables. Next, we present a completely new construction that itself starts from Maiorana-McFarland bent functions which are straightforward concatenations of linear functions., National Natural Science Foundation of China - 61872435 61602394 Fundamental Research Funds for the Central Universities - 2682016CX113 2682018ZT25

Two permutations of $[n]=\{1,2 \ldots n\}$ are \textit{$k$-neighbor separated} if there are two elements that are neighbors in one of the permutations and that are separated by exactly $k-2$ other elements in the other permutation. Let the maximal number of pairwise $k$-neighbor separated permutations of $[n]$ be denoted by $P(n,k)$. In a previous paper, the authors have determined $P(n,3)$ for every $n$, answering a question of K\"orner, Messuti and Simonyi affirmatively. In this paper we prove that for every fixed positive integer $\ell $, $$P(n,2^\ell+1) = 2^{n-o(n)}. $$ We conjecture that for every fixed even $k$, $P(n,k)=2^{n-o(n)}$. We also show that this conjecture is asymptotically true in the following sense $$\lim_{k \rightarrow \infty} \lim_{n \rightarrow \infty} \sqrt[n]{P(n,k)}=2.$$ Finally, we show that for even $n$, $P(n,n)= 3n/2$.

In this paper, we study zero-one laws for the Erd\H{o}s--R\'{e}nyi random graph model $G(n,p)$ in the case when $p = n^{-\alpha}$ for $\alpha>0$. For a given class $\mathcal{K}$ of logical sentences about graphs and a given function $p=p(n)$, we say that $G(n,p)$ obeys the zero-one law (w.r.t. the class $\mathcal{K}$) if each sentence $\varphi\in\mathcal{K}$ either a.a.s. true or a.a.s. false for $G(n,p)$. In this paper, we consider first order properties and monadic second order properties of bounded \textit{quantifier depth} $k$, that is, the length of the longest chain of nested quantifiers in the formula expressing the property. Zero-one laws for properties of quantifier depth $k$ we call the \textit{zero-one $k$-laws}. The main results of this paper concern the zero-one $k$-laws for monadic second order properties (MSO properties). We determine all values $\alpha>0$, for which the zero-one $3$-law for MSO properties does not hold. We also show that, in contrast to the case of the $3$-law, there are infinitely many values of $\alpha$ for which the zero-one $4$-law for MSO properties does not hold. To this end, we analyze the evolution of certain properties of $G(n,p)$ that may be of independent interest.

This is the last paper of a series of four papers in which we prove the following relaxation of the Loebl-Komlos-Sos Conjecture: For every $\alpha>0$ there exists a number~$k_0$ such that for every $k>k_0$ every $n$-vertex graph $G$ with at least $(\frac12+\alpha)n$ vertices of degree at least $(1+\alpha)k$ contains each tree $T$ of order $k$ as a subgraph. In the first two papers of this series, we decomposed the host graph $G$, and found a suitable combinatorial structure inside the decomposition. In the third paper, we refined this structure, and proved that any graph satisfying the conditions of the above approximate version of the Loebl-Komlos-Sos Conjecture contains one of ten specific configurations. In this paper we embed the tree $T$ in each of the ten configurations., Comment: 81 pages, 12 figures. A fix reflecting the change of Preconfiguration Clubs in Paper III, additional small changes

This is the third of a series of four papers in which we prove the following relaxation of the Loebl-Komlos-Sos Conjecture: For every $\alpha>0$ there exists a number $k_0$ such that for every $k>k_0$ every $n$-vertex graph $G$ with at least $(\frac12+\alpha)n$ vertices of degree at least $(1+\alpha)k$ contains each tree $T$ of order $k$ as a subgraph. In the first paper of the series, we gave a decomposition of the graph $G$ into several parts of different characteristics. In the second paper, we found a combinatorial structure inside the decomposition. In this paper, we will give a refinement of this structure. In the forthcoming fourth paper, the refined structure will be used for embedding the tree $T$., Comment: 59 pages, 4 figures; further comments by a referee incorporated; this includes a subtle but important fix to Lemma 5.1; as a consequence, Preconfiguration Clubs was changed

This is the second of a series of four papers in which we prove the following relaxation of the Loebl-Komlos--Sos Conjecture: For every $\alpha>0$ there exists a number $k_0$ such that for every $k>k_0$ every $n$-vertex graph $G$ with at least $(\frac12+\alpha)n$ vertices of degree at least $(1+\alpha)k$ contains each tree $T$ of order $k$ as a subgraph. In the first paper of the series, we gave a decomposition of the graph $G$ into several parts of different characteristics; this decomposition might be viewed as an analogue of a regular partition for sparse graphs. In the present paper, we find a combinatorial structure inside this decomposition. In the last two papers, we refine the structure and use it for embedding the tree $T$., Comment: 38 pages, 4 figures; new is Section 5.1.1; accepted to SIDMA

Assume that we are given a set of applicants and a set of posts such that each applicant has a preference list over the posts. A matching $M$ between the applicants and the posts is said to be popular if there is no other matching $N$ such that the number of applicants that prefer $N$ to $M$ is larger than the number of applicants that prefer $M$ to $N$. Then, the goal of the popular matching problem is to decide whether there is a popular matching, and find a popular matching if one exists. Abraham, Irving, Kavitha, and Mehlhorn proved that this problem can be solved in polynomial time even if the preference lists contain ties. In this paper, we consider the popular matching problem with matroid constraints. In this problem, for each post, we are given a matroid on the set of applicants. A set of applicants assigned to each post must be an independent set of its matroid. Kamiyama proved that if there is not a tie in the preference lists, then this problem can be solved in polynomial time. In this paper, ...

In this paper, lower bounds on the probability of a finite union of events are considered, i.e. $P\left(\bigcup_{i=1}^N A_i\right)$, in terms of the individual event probabilities $\{P(A_i), i=1,\ldots,N\}$ and the sums of the pairwise event probabilities, i.e., $\{\sum_{j:j\neq i} P(A_i\cap A_j), i=1,\ldots,N\}$. The contribution of this paper includes the following: (i) in the class of all lower bounds that are established in terms of only the $P(A_i)$'s and $\sum_{j:j\neq i} P(A_i\cap A_j)$'s, the optimal lower bound is given numerically by solving a linear programming (LP) problem with $N^2-N+1$ variables; (ii) a new analytical lower bound is proposed based on a relaxed LP problem, which is at least as good as the bound due to Kuai, et al.; (iii) numerical examples are provided to illustrate the performance of the bounds., Comment: SIAM Journal on Discrete Mathematics, 2016

Let ${{hom}}(G,H)$ denote the number of homomorphisms from a graph $G$ to a graph $H$. In this paper we study the number of homomorphisms of trees into a path, and prove that ${{hom}}(P_m,P_n)\leq {{hom}}(T_m,P_n)\leq {{hom}}(S_m,P_n),$ where $T_m$ is any tree on $m$ vertices, and $P_m$ and $S_m$ denote the path and star on $m$ vertices, respectively. This completes the study of extremal problems concerning the number of homomorphisms between trees started in the paper Graph Homomorphisms Between Trees [Electron. J. Combin., 21 (2014), 4.9] written by the authors of the current paper.

Extension of a graph $G$ is the construction of a new graph with certain properties by adding edges to some pairs of vertices in $G$. In this paper, we focus on extension of a quadrangulation of a surface to even triangulations, where a quadrangulation is a map on a surface with every face quadrangular and a triangulation is even if all the vertices have even degree. Zhang and He [SIAM J. Comput., 34 (2005), pp. 683--696] gave a formula for the exact number of distinct even triangulations extended from a given plane quadrangulation, and a lower bound of the number for the case of orientable nonspherical surfaces. They also posed the problem of finding the exact number for the latter case. In this paper, using topological methods, we improve the results by Zhang and He in the following directions: (I) extension of quadrangulations of a nonorientable surface and (II) complete enumeration of even triangulations extended from a given quadrangulation of a nonspherical surface. Indeed, we completely solve the p...

This paper answers an open question posed in 1993 concerning the least dense arrangement of a `full' packing of equal-sized disks. A disk $D$ in the packing is full if it touches at least four other disks and these disks leave no gap large enough for any other disk to touch $D$. A packing is called full if every disk is full. In the context of disks with unit diameter, Cowan [J. Appl. Probab., 30 (1993), pp. 263--268] showed that the intensity of disk centers (the average number per unit area) for any full packing must be greater than $12/(4+\sqrt3+5\tan(3\pi/10))= 0.95132\ldots$ and questioned if intensities less than $1$ were possible. In this paper, we answer his question affirmatively by showing that, for every intensity from $[0.95257\ldots,1]$, there exists a full packing of unit-diameter disks. We also establish that a full packing exists for all higher intensities up to the obvious maximum of $ \frac{2}{\sqrt{3}}$. Specific packings are constructed to demonstrate these existence issues.

In this paper, we investigate the permutation property of the Dickson polynomials $E_n(x, a)$ of the second kind over $\mathbb{Z}_m$. Due to a known result, it suffices to consider permutation polynomials $E_n(x, a)$ over $\mathbb{Z}_{p^t}$, where $p$ is a prime and $t$ is a positive integer. We identify all permutation polynomials of $E_n(x, a)$ over $\mathbb{Z}_{p^t}$ for (I) $p=2$ and (II) $p$ is odd and $a$ is a square over $\mathbb{Z}_p$. For odd $p$ and nonsquares $a$ in $\mathbb{Z}_p$, we determine a large class (if not all) of permutation polynomials $E_n(x, a)$ over $\mathbb{Z}_{p^t}$. A conjecture is also presented in this paper. If this conjecture is true, then all Dickson permutation polynomials $E_n(x, a)$ of the second kind over $\mathbb{Z}_m$ are determined.

Let $G$ be a graph with a threshold function $\theta:V(G)\rightarrow \mathbb{N}$ such that $1\leq \theta(v)\leq d_G(v)$ for every vertex $v$ of $G$, where $d_G(v)$ is the degree of $v$ in $G$. Suppose we are given a target set $S\subseteq V(G)$. The paper considers the following repetitive process on $G$. At time step 0 the vertices of $S$ are colored black and the other vertices are colored white. After that, at each time step $t>0$, the colors of white vertices (if any) are updated according to the following rule. All white vertices $v$ that have at least $\theta(v)$ black neighbors at the time step $t-1$ are colored black, and the colors of the other vertices do not change. The process runs until no more white vertices can update colors from white to black. The following optimization problem is called Target Set Selection: Finding a target set $S$ of smallest possible size such that all vertices in $G$ are black at the end of the process. Such an $S$ is called an {\em optimal target set} for $G$ under the threshold function $\theta$. We are interested in finding an optimal target set for the well-known class of honeycomb networks under an important threshold function called {\em strict majority threshold}, where $\theta(v)=\lceil (d_G(v)+1)/2\rceil$ for each vertex $v$ in $G$. In a graph $G$, a {\em feedback vertex set} is a subset $S\subseteq V(G)$ such that the subgraph induced by $V(G)\setminus S$ is cycle-free. In this paper we give exact value on the size of the optimal target set for various kinds of honeycomb networks under strict majority threshold, and as a by-product we also provide a minimum feedback vertex set for different kinds regular graphs in the class of honeycomb networks

In earlier papers we have developed an algebraic theory of discrete tomography. In those papers the structure of the functions $f: A \to \{0,1\}$ and $f: A \to \mathbb{Z}$ having given line sums in certain directions have been analyzed. Here $A$ was a block in $\mathbb{Z}^n$ with sides parallel to the axes. In the present paper we assume that there is noise in the measurements and (only) that $A$ is an arbitrary or convex finite set in $\mathbb{Z}^n$. We derive generalizations of earlier results. Furthermore we apply a method of Beck and Fiala to obtain results of he following type: if the line sums in $k$ directions of a function $h: A \to [0,1]$ are known, then there exists a function $f: A \to \{0,1\}$ such that its line sums differ by at most $k$ from the corresponding line sums of $h$., Comment: 16 pages

Cyclic codes are a subclass of linear codes and have applications in consumer electronics, data storage systems, and communication systems as they have efficient encoding and decoding algorithms. In this paper, some monomials and trinomials over finite fields are employed to construct a number of families of cyclic codes. Lower bounds on the minimum weight of some families of the cyclic codes are developed. The minimum weights of other families of the codes constructed in this paper are determined. The dimensions of the codes are flexible. Many of the codes presented in this paper are optimal or almost optimal in the sense that they meet some bounds on linear codes. Open problems regarding cyclic codes from monomials and trinomials are also presented.

Using the correspondence between abstract regular polytopes and string C-groups, in a recent paper [M. E. Fernandes, D. Leemans, and M. Mixer, J. Combin. Theory Ser. A, 119 (2012), pp. 42–56], we constructed an abstract regular polytope of rank r, for each $r \geq 3$, with automorphism group isomorphic to $A_{2r+3}$ when r is odd, and $A_{2r+1}$ when r is even. In this paper, the remaining cases are completed. It is shown that every group $A_n$, with n sufficiently large, acts on at least one abstract regular polytope of rank $\lfloor\frac{n-1}{2}\rfloor$. We conjecture that this is the highest possible rank for $n\geq 12$.

A seminal result of Reed et al. [Combinatorica, 16 (1996), pp. 535--554] says that a directed graph has either $k$ vertex-disjoint directed circuits or a set of at most $f(k)$ vertices meeting all directed circuits. This paper aims at generalizing their result to packing directed circuits through a prescribed set $S$ of vertices. Such a circuit is called an $S$-circuit. Even et al. [Algorithmica, 20 (1998), pp. 151--174] showed a fractional version of packing $S$-circuits. In this paper, we show that the fractionality can be bounded by at most one-fifth: Given an integer $k$ and a vertex subset $S$, whose size may not depend on $k$, we prove that either $G$ has a $1/5$-integral packing of $k$ disjoint $S$-circuits, i.e., each vertex appears in at most five of these $S$-circuits, or $G$ has a vertex set $X$ of order at most $f(k)$ (for some function $f$ of $k$) such that $G-X$ has no such circuit. We also give a fixed-parameter tractable approximation algorithm for finding a $1/5$-integral packing of $S$-c...

The minimum cost homomorphism problem has arisen as a natural and useful optimization problem in the study of graph (and digraph) coloring and homomorphisms: it unifies a number of other well studied optimization problems. It was shown by Gutin, Rafiey, and Yeo that the minimum cost problem for homomorphisms to a digraph H that admits a so-called extended Min- Max ordering is polynomial time solvable, and these authors conjectured that for all other digraphs H the problem is NP-complete. In a companion paper, we gave a forbidden structure characterization of digraphs that admit extended Min-Max orderings. In this paper, we apply this characterization to prove Gutin's conjecture.

A generalized Davenport-Schinzel sequence is one over a finite alphabet that excludes subsequences isomorphic to a fixed forbidden subsequence. The fundamental problem in this area is bounding the maximum length of such sequences. Following Klazar, we let $\mathrm{Ex}(\sigma,n)$ be the maximum length of a sequence over an $n$-letter alphabet excluding subsequences isomorphic to $\sigma$. It has been proved that for every $\sigma$, $\mathrm{Ex}(\sigma,n)$ is either linear or very close to linear. In particular it is $O(n2^{\alpha(n)^{O(1)}})$, where $\alpha$ is the inverse-Ackermann function and $O(1)$ depends on $\sigma$. In much the same way that the complete graphs $K_5$ and $K_{3,3}$ represent the minimal causes of nonplanarity, there must exist a set $\Phi_{Nonlin}$ of minimal nonlinear forbidden subsequences. Very little is known about the size or membership of $\Phi_{Nonlin}$. In this paper we construct an infinite antichain of nonlinear forbidden subsequences which, we argue, strongly supports the conjecture that $\Phi_{Nonlin}$ is itself infinite. Perhaps the most novel contribution of this paper is a succinct, humanly readable code for expressing the structure of forbidden subsequences.

Let $G$ be a simple, undirected, finite graph with vertex set $V(G)$ and edge set $E(G)$. A $k$-dimensional box is a Cartesian product of closed intervals $[a_1,b_1]\times [a_2,b_2]\times\cdots\times [a_k,b_k]$. The boxicity of $G$, box$(G)$, is the minimum integer $k$ such that $G$ can be represented as the intersection graph of $k$-dimensional boxes; i.e., each vertex is mapped to a $k$-dimensional box and two vertices are adjacent in $G$ if and only if their corresponding boxes intersect. Let $\mathcal{P}=(S,P)$ be a poset, where $S$ is the ground set and $P$ is a reflexive, antisymmetric and transitive binary relation on $S$. The dimension of $\mathcal{P}$, $\dim(\mathcal{P})$, is the minimum integer $t$ such that $P$ can be expressed as the intersection of $t$ total orders. Let $G_{\mathcal{P}}$ be the underlying comparability graph of $\mathcal{P}$; i.e., $S$ is the vertex set and two vertices are adjacent if and only if they are comparable in $\mathcal{P}$. It is a well-known fact that posets with the same underlying comparability graph have the same dimension. The first result of this paper links the dimension of a poset to the boxicity of its underlying comparability graph. In particular, we show that for any poset $\mathcal{P}$, box$(G_{\mathcal{P}})/(\chi(G_{\mathcal{P}})-1) \le \dim(\mathcal{P})\le 2\mbox{box}(G_{\mathcal{P}})$, where $\chi(G_{\mathcal{P}})$ is the chromatic number of $G_{\mathcal{P}}$ and $\chi(G_{\mathcal{P}})\ne1$. It immediately follows that if $\mathcal{P}$ is a height-$2$ poset, then box$(G_{\mathcal{P}})\le \dim(\mathcal{P})\le 2\mbox{box}(G_{\mathcal{P}})$ since the underlying comparability graph of a height-$2$ poset is a bipartite graph. The second result of the paper relates the boxicity of a graph $G$ with a natural partial order associated with the extended double cover of $G$, denoted as $G_c$: Note that $G_c$ is a bipartite graph with partite sets $A$ and $B$ which are copies of $V(G)$ such that, corresponding to every $u\in V(G)$, there are two vertices $u_A\in A$ and $u_B\in B$ and $\{u_A,v_B\}$ is an edge in $G_c$ if and only if either $u=v$ or $u$ is adjacent to $v$ in $G$. Let $\mathcal{P}_c$ be the natural height-$2$ poset associated with $G_c$ by making $A$ the set of minimal elements and $B$ the set of maximal elements. We show that $\frac{\mbox{box}(G)}{2} \le \dim(\mathcal{P}_c) \le 2\mbox{box}(G)+4$. These results have some immediate and significant consequences. The upper bound $\dim(\mathcal{P})\le 2\mbox{box}(G_\mathcal{P})$ allows us to derive hitherto unknown upper bounds for poset dimension such as $\dim(\mathcal{P})\le 2\mbox{ tree width }(G_{\mathcal{P}})+4$, since boxicity of any graph is known to be at most its tree width $+\; 2$. In the other direction, using the already known bounds for partial order dimension we get the following: (1) The boxicity of any graph with maximum degree $\Delta$ is $O(\Delta\log^2\Delta)$, which is an improvement over the best-known upper bound of $\Delta^2+2$. (2) There exist graphs with boxicity $\Omega(\Delta\log\Delta)$. This disproves a conjecture that the boxicity of a graph is $O(\Delta)$. (3) There exists no polynomial-time algorithm to approximate the boxicity of a bipartite graph on $n$ vertices with a factor of $O(n^{0.5-\epsilon})$ for any $\epsilon>0$ unless $NP=ZPP$.

Sphere packing problems have a rich history in both mathematics and physics; yet, relatively few analytical analyses of sphere packings exist, and answers to seemingly simple questions are unknown. Here, we present an analytical method for deriving all packings of n spheres in R^3 satisfying minimal rigidity constraints (>= 3 contacts per sphere and >= 3n-6 total contacts). We derive such packings for n = 9, (iii) the number of ground states (i.e. packings with the maximal number of contacts) oscillates with respect to n, (iv) for 10, The updates that have been made to version II are as follows: (i) some updates to the main text have been made, primarily correcting the number of packings reported in PRL, 103, 118303 and in version I of this paper for n = 9,10. (ii) The supplemental information for the paper has been included

A neighbor system, introduced in this paper, is a collection of integral vectors in $\mathbb{R}^{n}$ with some special structure. Such collections (slightly) generalize jump systems, which, in turn, generalize integral bisubmodular polyhedra, integral polymatroids, delta-matroids, matroids, and other structures. We show that neighbor systems provide a systematic and simple way to characterize these structures. One of our main results is a simple greedy algorithm for optimizing over (finite) neighbor systems starting from any feasible vector. The algorithm is (essentially) identical to the usual greedy algorithm on matroids and integral polymatroids when the starting vector is zero. But in all other cases, from matroids through jump systems, it appears to be a new greedy algorithm. We end the paper by introducing another structure, which is more general than neighbor systems, and indicate that essentially the same greedy algorithm also works for this structure.

This paper deals with cooperative games in which only certain coalitions are allowed to form. There have been previous models developed to confront the problem of nonfeasible coalitions. Games restricted by a communication graph are games in which the feasible coalitions are those that induce connected subgraphs. Another type of model is determined by the positions of the players in a so-called permission structure. In this paper, the restrictions to the cooperation are given by a combinatorial structure called an augmenting system which generalizes antimatroid structure and the system of connected subgraphs of a graph. Furthermore, the class of augmenting systems includes the conjunctive and disjunctive systems derived from a permission structure. The value α is a generalization of the Myerson value for games restricted by graphs and the Shapley value for games restricted by permission structures. The main results of the paper are the characterization of the value α for augmenting structures by using component efficiency, loop-null, and balanced contributions, and another characterization by consistency of this value. Furthermore, we implement a direct algorithm to compute this value by using the outputs of the original game. Ministerio de Educación - Unión Europea SEJ2006–00706 Junta de Andalucía FQM 237

The balanced decomposition number $f(G)$ of a graph $G$ was introduced by Fujita and Nakamigawa [Discr. Appl. Math., 156 (2008), pp. 3339-3344]. A balanced coloring of a graph $G$ is a coloring of some of the vertices of $G$ with two colors, such that there is the same number of vertices in each color. Then, $f(G)$ is the minimum integer $s$ with the following property: For any balanced coloring of $G$, there is a partition $V(G)=V_1\,\dot\cup\,\cdots\,\dot\cup\,V_r$ such that, for every $i$, $V_i$ induces a connected subgraph, $|V_i|\leq s$, and $V_i$ contains the same number of colored vertices in each color. Fujita and Nakamigawa studied the function $f(G)$ for many basic families of graphs, and demonstrated some applications. In this paper, we shall continue the study of the function $f(G)$. We give a characterization for noncomplete graphs $G$ of order $n$ which are $\lfloor\frac{n}{2}\rfloor$-connected, in view of the balanced decomposition number. We shall prove that a necessary and sufficient condition for such $\lfloor\frac{n}{2}\rfloor$-connected graphs $G$ is $f(G)=3$. We shall also determine $f(G)$ when $G$ is a complete multipartite graph, and when $G$ is a generalized $\Theta$-graph (i.e., a graph which is a subdivision of a multiple edge). Some applications will also be discussed. Further results about the balanced decomposition number also appear in two subsequent papers by Fujita and Liu.

In a recent paper, Duran et al. [J. Algorithms, 58 (2006), pp. 67-78] described an algorithm of complexity $O(n^2)$ for recognizing whether a graph $G$ with $n$ vertices and $m$ edges is a unit circular-arc (UCA) graph. Furthermore, the following open questions were posed in the above paper: (i) Is it possible to construct a UCA model for $G$ in polynomial time? (ii) Is it possible to construct a UCA model, whose extremes of the arcs correspond to integers of polynomial size? (iii) If (ii) is true, could such a model be constructed in polynomial time? In the present paper, we describe a characterization of UCA graphs, based on network circulations. The characterization leads to a different recognition algorithm and to answering these questions in the affirmative. We construct a UCA model whose extremes of the arcs correspond to integers of size $O(n)$. The proposed algorithms, for recognizing UCA graphs and constructing UCA models, have complexities $O(n+m)$. Furthermore, the complexities reduce to $O(n)$, if a proper circular-arc (PCA) model of $G$ is already given as the input, provided the extremes of the arcs are ordered. We remark that a PCA model of $G$ can be constructed in $O(n+m)$ time, using the algorithm by Deng, Hell, and Huang [SIAM J. Comput., 25 (1996), pp. 390-403]. Finally, we also describe a linear time algorithm for finding feasible circulations in networks with nonnegative lower capacities and unbounded upper capacities. Such an algorithm is employed in the model construction for UCA graphs.

We consider approximation algorithms for the metric labeling problem. This problem was introduced in a paper by Kleinberg and Tardos [J. ACM, 49 (2002), pp. 616--630] and captures many classification problems that arise in computer vision and related fields. They gave an O(log k log log k) approximation for the general case, where k is the number of labels, and a 2-approximation for the uniform metric case. (In fact, the bound for general metrics can be improved to O(log k) by the work of Fakcheroenphol, Rao, and Talwar [Proceedings of the 35th Annual ACM Symposium on Theory of Computing, 2003, pp. 448--455].) Subsequently, Gupta and Tardos [Proceedings of the 32nd Annual ACM Symposium on the Theory of Computing, 2000, pp. 652--658] gave a 4-approximation for the truncated linear metric, a metric motivated by practical applications to image restoration and visual correspondence. In this paper we introduce an integer programming formulation and show that the integrality gap of its linear relaxation either matches or improves the ratios known for several cases of the metric labeling problem studied until now, providing a unified approach to solving them. In particular, we show that the integrality gap of our linear programming (LP) formulation is bounded by O(log k) for a general k-point metric and 2 for the uniform metric, thus matching the known ratios. We also develop an algorithm based on our LP formulation that achieves a ratio of $2+\sqrt{2}\simeq 3.414$ for the truncated linear metric improving the earlier known ratio of 4. Our algorithm uses the fact that the integrality gap of the LP formulation is 1 on a linear metric.

In this paper we study the following question posed by H. S. Wilf: what is, asymptotically as $n\rightarrow \infty$, the probability that a randomly chosen part size in a random composition of an integer n has multiplicity m? More specifically, given positive integers n and m, suppose that a composition $\lambda$ of n is selected uniformly at random and then, out of the set of part sizes in $\lambda$, a part size j is chosen uniformly at random. Let $\P(A_n^{(m)})$ be the probability that j has multiplicity m. We show that for fixed m, $\P(A_n^{(m)}$) goes to 0 at the rate $1/\ln n$. A more careful analysis uncovers an unexpected result: $(\ln n)\P(A_n^{(m)})$ does not have a limit but instead oscillates around the value $1/m$ as $n\to\infty$. This work is a counterpart of a recent paper of Corteel, Pittel, Savage, and Wilf, who studied the same problem in the case of partitions rather than compositions.

Let ${\cal G}_d (n)$ be the set of d-regular simple graphs with n vertices, and for $G = (V,E)$ in ${\cal G}_d (n)$ let $f: V \to \{1,2, \ldots, n \}$ be a bijective coloring of the vertices of G. Also let $\alpha$ denote an interference parameter in $\{0,1, \ldots, n-1 \}$, and define the interference number of f with respect to G and $\alpha$ as the number of edges $\{u,v\}$ in $E$ for which $\min \{ | f(u) - f(v) |$, $n- | f(u) - f(v) | \} \le \alpha$. We consider two interference number problems for feasible $(n, \alpha, d)$. The first is to specify a $G \in {\cal G}_d (n)$ and an f for which the interference number is as small as possible. The second is to determine a $G \in {\cal G}_d (n)$ whose minimum interference number over all f is as large as possible. A previous paper completely solves both problems for d = 2. The present paper solves the first problem for all $d \ge 3$ and obtains partial results for the second problem for $d \ge 3$ that focus on interference-minimizing f's when G consists of disjoint copies of $K_{d+1}$.

In this paper we show that many general lower bounds on the broadcasting and gossiping time are optimal. In particular, let b(G) be the broadcasting time of a network G under the basic one-port model. The only lower bound on b(G) holding for every n vertices graph G is max$(\log_2 n, Diam(G))$, but the $\log_2 n$ factor cannot be achieved in bounded degree networks. In fact, let the parameter d be defined in undirected graphs as the maximum degree minus one and for directed graphs as the maximum out-degree. Then, in [ SIAM J. Discrete Math., 1 (1998), pp. 531--540; SIAM J. Discrete Math., 5 (1992), pp. 10--24] it has been proved that, for any graph G of parameter d, b(G) \geq \frac{\log_2 n}{\log_2 \xi}$, where $\xi$ is the largest real number such that $\xi^{d} -\xi^{d-1} - \xi^{d-2} -\cdots - \xi-1=0$. Since then many papers have proposed constructions of bounded degree networks having a small broadcast time [Proceedings of the 2nd International Euro-Par Conference (EUROPAR), Lecture Notes in Comput. Sci. 1123, Springer-Verlag, New York, 1996, pp. 313--324; IEEE Trans. Comput., 33 (1984), pp. 190--194], but so far the optimality of [SIAM J. Discrete Math., 1 (1998), pp. 531--540; {SIAM J. Discrete Math.}, 5 (1992), pp. 10--24] was still an open question. In this paper we prove that the above lower bound is tight, improving all the existing upper bounds by means of probabilistic methods. Namely, we show that for n arbitrarily large there exist families of n vertices graphs in which a uniformly drawn graph has broadcasting time as predicted by [SIAM J. Discrete Math., 1 (1998), pp. 531--540; SIAM J. Discrete Math., 5 (1992), pp. 10--24] with probability converging to 1. Moreover, we show that [SIAM J. Discrete Math., 1 (1998), pp. 531--540; SIAM J. Discrete Math., 5 (1992), pp. 10--24] is attained even in the case of gossiping and systolic gossiping in the full-duplex mode. Finally, new upper bounds on bounded-degree and systolic gossiping are also determined in the directed and half-duplex modes. While the systolic construction is tight and matches the lower bound of [Inform and Comput., to appear], we strongly conjecture that the bounded-degree result is optimal and that a corresponding matching lower bound is still to be proven.

Given a directed graph $G$ and an arc weight function $w: E(G)\rightarrow\mathbb{R}_+$, the maximum directed cut problem ({\sc max dicut}) is that of finding a directed cut $\delta (X)$ with maximum total weight. In this paper we consider a version of {\sc max dicut}---{\sc max dicut} with given sizes of parts or {\sc max dicut with gsp}---whose instance is that of {\sc max dicut} plus a positive integer $p$, and it is required to find a directed cut $\delta (X)$ having maximum weight over all cuts $\delta (X)$ with $|X|=p$. Our main result is a $0.5$-approximation algorithm for solving the problem. The algorithm is based on a tricky application of the pipage rounding technique developed in some earlier papers by two of the authors and a remarkable structural property of basic solutions to a linear relaxation. The property is that each component of any basic solution is an element of a set $\{0,\delta,1/2,1-\delta,1 \}$, where $\delta$ is a constant that satisfies $0

The graph of linear extensions G(P) of a poset P has as vertices the linear extensions of P, and two vertices are adjacent if they differ only by an adjacent transposition. This graph has so far been investigated mainly with respect to Hamilton paths and intrinsic geodesic convexity. The work on the latter topic especially shows how the graph-theoretic properties of G(P) reflect the order-theoretic structure of P. The aim of this paper is to study the graph G(P) with respect to topics from classical graph theory, e.g., connectivity, cycle space, isometric embeddings, and to find order-theoretic interpretations of these notions. The main theorems in this paper are the equality of the connectivity of G(P) and the jump number of P, the existence of a certain generating system for the cycle space, and a relationship of P and its subposets obtained via an embedding of the graph of linear extensions.

This is the first half of a two-part paper devoted to on-line 3-colorable graphs. Here on-line 3-colorable triangle-free graphs are characterized by a finite list of forbidden induced subgraphs. The key role in our approach is played by the family of graphs which are both triangle- and (2K2 + K1)-free. Characterization of this family is given by introducing a bipartite modular decomposition concept. This decomposition, combined with the greedy algorithm, culminates in an on-line 3-coloring algorithm for this family. On the other hand, based on the characterization of this family, all 22 forbidden subgraphs of on-line 3-colorable triangle-free graphs are determined. As a corollary, we obtain the 10 forbidden subgraphs of on-line 3-colorable bipartite graphs. The forbidden subgraphs in the finite basis characterization are on-line 4-critical, i.e., they are on-line 4-chromatic but their proper induced subgraphs are on-line 3-colorable. The results of this paper are applied in the companion paper [Discrete Math., 177 (1997), pp. 99--122] to obtain the finite basis characterization of connected on-line 3-colorable graphs (with 51 4-critical subgraphs). However, perhaps surprisingly, connectivity (or the triangle-free property) is essential in a finite basis characterization: there are infinitely many on-line 4-critical graphs.

This paper studies the problem of embedding a graph G into a book with vertices on a line along the spine of the book and edges on the pages in such a way that no edge crosses another. When each edge is allowed to appear in one or more pages by crossing the spine, one of the authors showed that there exists a three page book embedding of G in which each edge crosses the spine at most O(n) times, where n is the number of vertices. This paper improves the result and shows that there exists a three page book embedding of G in which each edge crosses the spine at most O(log n) times. An $\Omega(n^2)$ lower bound on the number of crossings of edges over the spine in any book embedding of the complete graph Kn is also shown.

Recently in several papers, graphs with maximum neighborhood orderings were characterized and turned out to be algorithmically useful. This paper gives a unified framework for characterizations of those graphs in terms of neighborhood and clique hypergraphs which have the Helly property and whose line graph is chordal. These graphs are dual (in the sense of hypergraphs) to chordal graphs. By using the hypergraph approach in a systematical way new results are obtained, some of the old results are generalized, and some of the proofs are simplified.

This is a sequel to a previous paper entitled The Order Dimension of Convex Polytopes, by the same authors [SIAM J. Discrete Math., 6 (1993), pp. 230--245]. In that paper, we considered the poset {{\bf P}\protect\boldmath$_M\!\!\!$} formed by taking the vertices, edges, and faces of a 3-connected planar map {\bf M}, ordered by inclusion, and showed that the order dimension of {{\bf P}\protect\boldmath$_M\!\!\!$} is always equal to 4. In this paper, we show that if {\bf M} is any planar map, then the order dimension of {{\bf P}\protect\boldmath$_M\!\!\!$} is still at most 4.

Let $\lambda({\cal N})$ denote the weight of a minimum cut in an edge-weighted undirected network ${\cal N}$, and $n$ and $m$ denote the numbers of vertices and edges, respectively. It is known that $O(n^{2k})$ is an upper bound on the number of cuts with weights less than $k\lambda({\cal N})$, where $k\geq 1$ is a given constant. This paper first shows that all cuts of weights less than $k\lambda({\cal N})$ can be enumerated in $O(m^2n+n^{2k}m)$ time without using the maximum flow algorithm. The paper then proves for $k

This paper considers counting problems associated with K-spanning and K-disconnecting sets for a specified terminal set K in an undirected graph G. In particular, we consider the problems of computing the number of Steiner trees and min K-cuts for G, as well as K-spanning and K-disconnecting sets of cardinality close to the minimum values. Among other things, these numbers are critical to the efficient approximation of K-connected reliability measures in stochastic networks. Although the counting problems considered in this paper are NP-hard in general, a large number of methods for finding shortest paths, min cuts, and Steiner trees in graphs can be extended to efficiently count K-spanning and K-disconnecting sets in important special cases.

Orthogonal arrays (OAs) are basic combinatorial structures, which appear under various disguises in cryptology and the theory of algorithms. Among their applications are universal hashing, authentication codes, resilient and correlation-immune functions, derandomization of algorithms, and perfect local randomizers. In this paper, we give new explicit bounds on the size of orthogonal arrays using Delsarte's linear programming method. Specifically, we prove that the minimum number of rows in a binary orthogonal array of length $n$ and strength $t$ is at least $ 2^{n} - (n 2^{n-1}/t+1)$ and also at least $ 2^{n} - (2^{n-2}(n+1)/\lceil \frac{t+1}{2} \rceil).$ We also prove that these bounds are as powerful as the linear programming bound itself for many parametric situations. An $(n,m,t)$-resilient function is a function $f: \{0,1\}^{n} \longrightarrow \{0,1\}^{m}$ such that every possible output $m$-tuple is equally likely to occur when the values of $t$ arbitrary inputs are fixed by an opponent and the remaining $n-t$ input bits are chosen independently at random. A basic problem is to maximize $t$ given $m$ and $n$, i.e., to determine the largest value of $t$ such that an $(n,m,t)$-resilient function exists. In this paper, we obtain upper and lower bounds for the optimal values of $t$ where $1 \leq n \leq 25$ and $1 \leq m < n$. The upper bounds are derived from Delsarte's linear programming bound, and the lower bounds come from constructions based on error-correcting codes. We also obtain new explicit upper bounds for the optimal values of $t$. It was proved by Chor et al. in [{\em Proc. {\rm 26}th IEEE Symp. on Foundations of Computer Science}, 1985, pp. 396--407] that an $(n,2,t)$-resilient function exists if and only if $ t < \lfloor \frac{2n}{3} \rfloor$. This result was generalized by Friedman [{\em Proc. $33$rd IEEE Symp. on Foundations of Computer Science}, 1992, pp. 314--319], who proved a bound for general $m$. We also prove some new bounds, and complete the determination of the optimal resiliency of resilient functions with $m=3$ and most of the cases for $m=4$. Several other infinite classes of (optimal) resilient functions are also constructed using the theory of anticodes.

To protect sensitive information in a cross tabulated table, it is a common practice to suppress some of the cells in the table. This paper investigates four levels of data security of a two-dimensional table concerning the effectiveness of this practice. These four levels of data security protect the information contained in, respectively, individual cells, individual rows and columns, several rows or columns as a whole, and a table as a whole. The paper presents efficient algorithms and NP-completeness results for testing and achieving these four levels of data security. All these complexity results are obtained by means of fundamental equivalences between the four levels of data security of a table and four types of connectivity of a graph constructed from that table.

This paper provides an efficient method to find all feasible offsets for a given separation in a very large-scale integration (VLSI) channel-routing problem in one layer. The previous literature considers this task only for problems with no single-sided nets. When single-sided nets are included, the worst-case solution time increases from $\Theta(n)$ to $\Omega(n^2)$, where $n$ is the number of nets. But if the number of columns $c$ is $O(n)$, the problem can be solved in time $O(n^{1.5}\lgn)$, which improves upon a "naive" $O(cn)$ approach. As a corollary of this result, the same time bound suffices to find the optimal offset (the one that minimizes separation). Better running times result when there are no two-sided nets or all single-sided nets are on one side of the channel. This paper also gives improvements upon the naive approach for $c\neq O(n)$, including an algorithm with running time independent of $c$. An interesting algorithmic aspect of the paper is a connection to discrete convolution.

This paper deals with the problem of broadcasting in minimum time in the telephone and message-passing models. Approximation algorithms are developed for arbitrary graphs as well as for several restricted graph classes. In particular, an $O(\sqrt{n})$-additive approximation algorithm is given for broadcasting in general graphs, and an $O(\log n/\log\log n)$ (multiplicative) ratio approximation is given for broadcasting in the open-path model. This also results in an algorithm for broadcasting on random graphs (in the telephone and message-passing model) that yields an $O(\log n/\log\log n)$ approximation with high probability. In addition, the paper presents a broadcast algorithm for graph families with small separators (such as chordal, $k$-outerplanar, bounded-face planar, and series-parallel graphs), with approximation ratio proportional to the separator size times $\log n$. Finally, an efficient approximation algorithm is presented for the class of graphs representable as trees of cliques.

Let $\cay(m,A)$ denote the Cayley digraph on ${\bf Z}_m$ generated by $A$, where ${\bf Z}_m$ is the cyclic group of residues modulo $m$. Let $r(m,A)$ denote the average distance of $\cay(m,A)$. For any $r\ge1$ and $k\ge1$ define $m^*(r,k)$ as the largest positive integer $m$ such that the average distance of the Cayley digraph $\hbox{Cay}(m,A)$ is at most $r$ for some set $A$ with $k$ elements. In this paper, an asymptotic formula for $m^*(r,2)$ is proved and a lower bound for $m^*(r,k)$ is also obtained for $k\ge3$. Applications to the construction of optimal distributed loop networks is discussed in this paper. A lower bound of the average order of subsets for asymptotic bases in number theory is proved using the main theorem of this paper.

A plane multigraph is said to be dual Eulerian if both itself and its dual contain an Euler path or circuit and the Euler paths have corresponding edge sequences. In this paper several properties of plane multigraphs are derived, and a necessary and sufficient condition for a plane multigraph to be dual Eulerian is given. Although the necessary and sufficient condition for a multigraph to be Eulerian is somewhat trivial, the necessary and sufficient condition for a plane multigraph to be dual Eulerian is not. Nevertheless, the question of whether or not a plane multigraph is dual Eulerian can be answered in time proportional to a linear function of the number of edges of the graph, and an algorithm that answers this question is presented in this paper. This theory can be applied to the layout synthesis of functional cells for Complementary Metal-Oxide Semiconductor Very Large-Scale Integrated circuits.

This paper examines the class of vertex-colored graphs that can be triangulated without the introduction of edges between vertices of the same color. This is related to a fundamental and long-standing problem for numerical taxonomists, called the Perfect Phylogeny Problem. These problems are known to be polynomially equivalent and NP-complete. This paper presents a dynamic programming algorithm that can be used to determine whether a given vertex-colored graph can be so triangulated and that runs in $O((n+m(k-2))^{k+1})$ time, where the graph has $n$ vertices, $m$ edges, and $k$ colors. The corresponding algorithm for the Perfect Phylogeny Problem runs in $O(r^{k+1} k^{k+1} + sk^2 )$ time, where $s$ species are defined by $k$ $r$-state characters.

This paper investigates a transformation P -> Q between partial orders P, Q that transforms the interval dimension of P to the dimension of Q, i.e., idim (P) = dim (Q). Such a construction has been shown before in the context of Ferrer's dimension by Cogis [Discrete Math., 38 (1982), pp. 47-52 ]. The construction in this paper can be shown to be equivalent to his, but it has the advantage of (1) being purely order-theoretic, (2) providing a geometric interpretation of interval dimension similar to that of Ore [Amer. Math. Soc. Colloq. Publ., Vol. 38, 1962] for dimension, and (3) revealing several somewhat surprising connections to other order-theoretic results. For instance, the transformation P -> Q can be seen as almost an inverse of the well-known split operation; it provides a theoretical background for the influence of edge subdivision on dimension (e.g., the results of Spinrad [Order, 5 (1989), pp. 143-147]) and interval dimension, and it turns out to be invariant with respect to changes of P that do not alter its comparability graph, thus also providing a simple new proof for the comparability invariance of interval dimension.