On the number of connected convex subgraphs of a connected acyclic digraph

A digraph D is connected if the underlying undirected graph of D is connected. A subgraph H of an acyclic digraph D is convex if there is no directed path between vertices of H which contains an arc not in H. We find the minimum and maximum possible number of connected convex subgraphs in a connected acyclic digraph of order n. Connected convex subgraphs of connected acyclic digraphs are of interest in the area of modern embedded processors technology.     

Acyclically pushable bipartite permutation digraphs: An algorithm

Given a digraph D=(V,A) and an X⊆V, D^X denotes the digraph obtained from D by reversing those arcs with exactly one end in X. A digraph D is called acyclically pushable when there exists an X⊆V such that D^X is acyclic. Huang, MacGillivray and Yeo have recently characterized, in terms of two excluded induced subgraphs on 7 and 8 nodes, those bipartite permutation digraphs which are acyclically pushable. We give an algorithmic proof of their result. Our proof delivers an O(m²) time algorithm to decide whether a bipartite permutation digraph is acyclically pushable and, if yes, to find a set X such that D^X is acyclic. (Huang, MacGillivray and Yeo's result clearly implies an O(n⁸) time algorithm to decide but the polynomiality of constructing X was still open.)We define a strongly acyclic digraph as a digraph D such that D^X is acyclic for every X. We show how a result of Conforti et al [Balanced cycles and holes in bipartite graphs, Discrete Math. 199 (1-3) (1999) 27-33] can be essentially regarded as a characterization of strongly acyclic digraphs and also provides linear time algorithms to find a strongly acyclic orientation of an undirected graph, if one exists. Besides revealing this connection, we add simplicity to the structural and algorithmic results first given in Conforti et al [Balanced cycles and holes in bipartite graphs, Discrete Math. 199 (1-3) (1999) 27-33]. In particular, we avoid decomposing the graph into triconnected components.We give an alternate proof of a theorem of Huang, MacGillivray and Wood characterizing acyclically pushable bipartite tournaments. Our proof leads to a linear time algorithm which, given a bipartite tournament as input, either returns a set X such that D^X is acyclic or a proof that D is not acyclically pushable.     

Primitive digraphs with smallest large exponent

A primitive digraph D on n vertices has large exponent if its exponent, γ(D), satisfies α_n?γ(D)?w_n, where α_n=⌊w_n/2⌋+2 and w_n=(n-1)²+1. It is shown that the minimum number of arcs in a primitive digraph D on n?5 vertices with exponent equal to α_n is either n+1 or n+2. Explicit constructions are given for fixed n even and odd, for a primitive digraph on n vertices with exponent α_n and n+2 arcs. These constructions extend to digraphs with some exponents between α_n and w_n. A necessary and sufficient condition is presented for the existence of a primitive digraph on n vertices with exponent α_n and n+1 arcs. Together with some number theoretic results, this gives an algorithm that determines for fixed n whether the minimum number of arcs is n+1 or n+2.     

Spectral radius of digraphs with given dichromatic number

Let D be a digraph with vertex set V(D). A partition of V(D) into k acyclic sets is called a k-coloring of D. The minimum integer k for which there exists a k-coloring of D is the dichromatic number χ(D) of the digraph D. Denote G_n,k the set of the digraphs of order n with the dichromatic number k≥2. In this note, we characterize the digraph which has the maximal spectral radius in G_n,k. Our result generalizes the result of [8] by Feng et al.     

Some sufficient conditions for the existence of kernels in infinite digraphs

A kernel N of a digraph D is an independent set of vertices of D such that for every w∈V(D)−N there exists an arc from w to N. If every induced subdigraph of D has a kernel, D is said to be a kernel perfect digraph. D is called a critical kernel imperfect digraph when D has no kernel but every proper induced subdigraph of D has a kernel. If F is a set of arcs of D, a semikernel modulo F of D is an independent set of vertices S of D such that for every z∈V(D)−S for which there exists an (S,z)-arc of D−F, there also exists an (z,S)-arc in D. In this work we show sufficient conditions for an infinite digraph to be a kernel perfect digraph, in terms of semikernel modulo F. As a consequence it is proved that symmetric infinite digraphs and bipartite infinite digraphs are kernel perfect digraphs. Also we give sufficient conditions for the following classes of infinite digraphs to be kernel perfect digraphs: transitive digraphs, quasi-transitive digraphs, right (or left)-pretransitive digraphs, the union of two right (or left)-pretransitive digraphs, the union of a right-pretransitive digraph with a left-pretransitive digraph, the union of two transitive digraphs, locally semicomplete digraphs and outward locally finite digraphs.     

Digraphs with real and gaussian spectra

The conventional binary operations of cartesian product, conjunction, and composition of two digraphs D₁ and D₂ are observed to give the sum, the product, and a more complicated combination of the spectra of D₁ and D₂ as the resulting spectrum. These formulas for analyzing the spectrum of a digraph are utilized to construct for any positive integer n, a collection of n nonisomorphic strong regular nonsymmetric digraphs with real spectra. Further, an infinite collection of strong nonsymmetric digraphs with nonzero gaussian integer value is found. Finally, for any n, it is shown that there are n cospectral strong nonsymmetric digraphs with integral spectra.     

Hypotraceable digraphs

A hypotraceable digraph is a digraph D = (V, E) which is not traceable, i.e., does not contain a (directed)Hamiltonian path, but for which D - v is traceable for all ve ∈ V. We prove that a hypotraceable digraph of order n exists iff n ≥ 7 and that for each k ≥ 3 there are infinitely many hypotraceable oriented graphs with a source and a sink and precisely k strong components. We also show that there are strongly connected hypotraceable oriented graphs and that there are hypotraceable digraphs with precisely two strong components one of which is a source or a sink. Finally, we prove that hypo-Hamiltonian and hypotraceable digraphs may contain large complete subdigraphs.     

A lower bound for the spectral radius of a digraph

We show that the spectral radius ρ(D) of a digraph D with n vertices and c₂ closed walks of length 2 satisfies Moreover, equality occurs if and only if D is the symmetric digraph associated to a -regular graph, plus some arcs that do not belong to cycles. As an application of this result, we construct new sharp upper bounds for the low energy of a digraph, which extends Koolen and Moulton bounds of the energy to digraphs.     

Linear arboricity of regular digraphs

A linear directed forest is a directed graph in which every component is a directed path.The linear arboricity la(D) of a digraph D is the minimum number of linear directed forests in D whose union covers all arcs of D. For every d-regular digraph D, Nakayama and P′eroche conjecture that la(D) = d + 1. In this paper, we consider the linear arboricity for complete symmetric digraphs,regular digraphs with high directed girth and random regular digraphs and we improve some wellknown results. Moreover, we propose a more precise conjecture about the linear arboricity for regular digraphs.     

Finding cheapest cycles in vertex-weighted quasi-transitive and extended semicomplete digraphs

We consider the problem of finding a minimum cost cycle in a digraph with real-valued costs on the vertices. This problem generalizes the problem of finding a longest cycle and hence is NP-hard for general digraphs. We prove that the problem is solvable in polynomial time for extended semicomplete digraphs and for quasi-transitive digraphs, thereby generalizing a number of previous results on these classes. As a byproduct of our method we develop polynomial algorithms for the following problem: Given a quasi-transitive digraph D with real-valued vertex costs, find, for each j=1,2,…,|V(D)|, j disjoint paths P₁,P₂,…,P_j such that the total cost of these paths is minimum among all collections of j disjoint paths in D.     

Kernels in edge-coloured orientations of nearly complete graphs

We call the digraph D an orientation of a graph G if D is obtained from G by the orientation of each edge of G in exactly one of the two possible directions. The digraph D is an m-coloured digraph if the arcs of D are coloured with m-colours.Let D be an m-coloured digraph. A directed path (or a directed cycle) is called monochromatic if all of its arcs are coloured alike.A set N⊆V(D) is said to be a kernel by monochromatic paths if it satisfies the two following conditions: (i) for every pair of different vertices u,v∈N there is no monochromatic directed path between them and (ii) for every vertex x∈V(D)-N there is a vertex y∈N such that there is an xy-monochromatic directed path.In this paper we obtain sufficient conditions for an m-coloured orientation of a graph obtained from K_n by deletion of the arcs of K_1,r(0?r?n-1) to have a kernel by monochromatic.     

Multiplicativity of acyclic local tournaments

A homomorphism of a digraph to another digraph is an edge-preserving vertex mapping. A digraphH is said to be multiplicative if the set of digraphs which do not admit a homomorphism toH is closed under categorical product. In this paper we discuss the multiplicativity of acyclic Hamiltonian digraphs, i.e., acyclic digraphs which contains a Hamiltonian path. As a consequence, we give a complete characterization of acyclic local tournaments with respect to multiplicativity.     

About some cyclic properties in digraphs

In this article we are concerned with digraphs in which any two vertices are on a common cycle. For example, we prove that, in a strong digraph of order n and half degrees at least 2 with at least n² ? 5n + 15 arcs, any two vertices are on a common cycle. We also consider related properties and give sufficient conditions on half degrees and the number of arcs to insure these properties. In particular, we show that every digraph of order n with half degrees at least r and with at least n² ? rn + r² arcs is 2-linked.     

Minimal n-fach zusammenhängende Digraphen

For a minimally n-connected digraph D, the subgraph spanned by the edges (x, y) with outdegree of x and indegree of y exceeding n is denoted by D₀. It is proved that D₀ corresponds to a forest. This implies that in a finite, minimally n-connected digraph D, the number of vertices of outdegree n is at least n and that the number of vertices of outdegree or indegree equal to n grows linearly in |D|. For almost every integer m, the maximum number of edges in a minimally n-connected digraph of order m is determined and the extremal digraphs are characterized.     

Relative perturbation bounds for the unitary polar factor

LetB be anm×n (m≥n) complex (or real) matrix. It is known that there is a uniquepolar decomposition B=QH, whereQ*Q=I, then×n identity matrix, andH is positive definite, providedB has full column rank. Existing perturbation bounds suggest that in the worst case, for complex matrices the change inQ be proportional to the reciprocal ofB's least singular value, or the reciprocal of the sum ofB's least and second least singular values if matrices are real. However, there are situations where this unitary polar factor is much more accurately determined by the data than the existing perturbation bounds would indicate. In this paper the following question is addressed: how much mayQ change ifB is perturbed to $\tilde B = D_1^* BD_2 $ , whereD ₁ andD ₂ are nonsingular and close to the identity matrices of suitable dimensions? It is shown that for a such kind of perturbation, the change inQ is bounded only by the distances fromD ₁ andD ₂ to identity matrices and thus is independent ofB's singular values. Such perturbation is restrictive, but not unrealistic. We show how a frequently used scaling technique yields such a perturbation and thus scaling may result in better-conditioned polar decompositions.     

Turán theorems and convexity invariants for directed graphs

This paper is motivated by the desire to evaluate certain classical convexity invariants (specifically, the Helly and Radon numbers) in the context of transitive closure of arcs in the complete digraph. To do so, it is necessary to establish several new Turán type results for digraphs and characterize the associated extremal digraphs. In the case of the Radon number, we establish the following analogue for transitive closure in digraphs of Radon's classical convexity theorem [J. Radon, Mengen konvexer Körper, die einer gemeinsamen Punkt enthalten, Math. Ann. 83 (1921) 113-115]: in a complete digraph on n?7 vertices with >n²/4 arcs, it is possible to partition the arc set into two subsets whose transitive closures have an arc in common.     

Irregularity strength of digraphs

It is an elementary exercise to show that any non-trivial simple graph has two vertices with the same degree. This is not the case for digraphs and multigraphs. We consider generating irregular digraphs from arbitrary digraphs by adding multiple arcs. To this end, we define an irregular labeling of a digraph D to be an arc-labeling of the digraph such that the ordered pairs of the sums of the in-labels and out-labels at each vertex are all distinct. We define the strength of D to be the smallest of the maximum labels used across all irregular labelings. Similar definitions for graphs have been studied extensively and a different formulation of digraph irregularity was given in [H. Hackett, Irregularity strength of graphs and digraphs, Masters Thesis, University of Louisville, 1995]. Here we continue the study of irregular labelings of digraphs. We give a general lower bound on and determine exactly for tournaments, directed paths and cycles and the orientation of the path where all vertices have either in-degree 0 or out-degree 0. We also determine the irregularity strength of a union of directed cycles and a union of directed paths, the latter which requires a new result pertaining to finding circuits of given lengths containing prescribed vertices in the complete symmetric digraph with loops.     

The diameter of randomly perturbed digraphs and some applications

The central observation of this paper is that if εn random arcs are added to any n‐node strongly connected digraph with bounded degree then the resulting graph has diameter 𝒪(lnn) with high probability. We apply this to smoothed analysis of algorithms and property testing. Smoothed Analysis: Recognizing strongly connected digraphs is a basic computational task in graph theory. Even for digraphs with bounded degree, it is NL‐complete. By XORing an arbitrary bounded degree digraph with a sparse random digraph R ∼ 𝔻_n,ε/n we obtain a “smoothed” instance. We show that, with high probability, a log‐space algorithm will correctly determine if a smoothed instance is strongly connected. We also show that if NL ⫅̸ almost‐L then no heuristic can recognize similarly perturbed instances of (s,t)‐connectivity. Property Testing: A digraph is called k‐linked if, for every choice of 2k distinct vertices s₁,…,s_k,t₁,…,t_k, the graph contains k vertex disjoint paths joining s_r to t_r for r = 1,…,k. Recognizing k‐linked digraphs is NP‐complete for k ≥ 2. We describe a polynomial time algorithm for bounded degree digraphs, which accepts k‐linked graphs with high probability, and rejects all graphs that are at least εn arcs away from being k‐linked. © 2007 Wiley Periodicals, Inc. Random Struct. Alg., 2007     

On semicomplete multipartite digraphs whose king sets are semicomplete digraphs

Reid [Every vertex a king, Discrete Math. 38 (1982) 93-98] showed that a non-trivial tournament H is contained in a tournament whose 2-kings are exactly the vertices of H if and only if H contains no transmitter. Let T be a semicomplete multipartite digraph with no transmitters and let K_r(T) denote the set of r-kings of T. Let Q be the subdigraph of T induced by K₄(T). Very recently, Tan [On the kings and kings-of-kings in semicomplete multipartite digraphs, Discrete Math. 290 (2005) 249-258] proved that Q contains no transmitters and gave an example to show that the direct extension of Reid's result to semicomplete multipartite digraphs with 2-kings replaced by 4-kings is not true. In this paper, we (1) characterize all semicomplete digraphs D which are contained in a semicomplete multipartite digraph whose 4-kings are exactly the vertices of D. While it is trivial that K₄(Q)⊆K₄(T), Tan [On the kings and kings-of-kings in semicomplete multipartite digraphs, Discrete Math. 290 (2005) 249-258] showed that K₃(Q)⊆K₃(T) and K₂(Q)=K₂(T). Tan [On the kings and kings-of-kings in semicomplete multipartite digraphs, Discrete Math. 290 (2005) 249-258] also provided an example to show that K₃(Q) need not be the same as K₃(T) in general and posed the problem: characterize all those semicomplete multipartite digraphs T such that K₃(Q)=K₃(T). In the course of proving our result (1), we (2) show that K₃(Q)=K₃(T) for all semicomplete multipartite digraphs T with no transmitters such that Q is a semicomplete digraph.     

The minimum spanning strong subdigraph problem is fixed parameter tractable

A digraph D is strong if it contains a directed path from x to y for every choice of vertices x,y in D. We consider the problem (MSSS) of finding the minimum number of arcs in a spanning strong subdigraph of a strong digraph. It is easy to see that every strong digraph D on n vertices contains a spanning strong subdigraph on at most 2n−2 arcs. By reformulating the MSSS problem into the equivalent problem of finding the largest positive integer k≤n−2 so that D contains a spanning strong subdigraph with at most 2n−2−k arcs, we obtain a problem which we prove is fixed parameter tractable. Namely, we prove that there exists an O(f(k)n^c) algorithm for deciding whether a given strong digraph D on n vertices contains a spanning strong subdigraph with at most 2n−2−k arcs.We furthermore prove that if k≥1 and D has no cut vertex then it has a kernel of order at most (2k−1)². We finally discuss related problems and conjectures.