The Evolution of Random Subgraphs of the Cube

be a random Qⁿ”-process, that is let Q₀ be the empty spanning subgraph of the cube Qⁿ and, for 1 ? t ? M = nN/2 = n2^n?1, let the graph Q_t be obtained from Q_t?1 by the random addition of an edge of Q_n not present in Q_t?1. When t is about N/2, a typical Q_t undergoes a certain “phase transition'': the component structure changes in a sudden and surprising way. Let t = (1 + ?) N/2 where ? is independent of n. Then all the components of a typical Q_t have o(N) vertices if ? < 0, while if ? > 0 then, as proved by Ajtai, Komlós, and Szemerédi, a typical Q_t has a “giant” component with at least α(?)N vertices, where α(?) > 0. In this note we give essentially best possible results concerning the emergence of this giant component close to the time of phase transition. Our results imply that if η > 0 is fixed and t ? (1 ? n^?η) N/2, then all components of a typical Q_t have at most n^β(η) vertices, where β(η) > 0. More importantly, if 60(log n)³/n ? ? = ?n = o(1), then the largest component of a typical Q_t has about 2?N vertices, while the second largest component has order O(n?^?2). Loosely put, the evolution of a typical Qⁿ process is such that shortly after time N/2 the appearance of each new edge results in the giant component acquiring 4 new vertices.     

Random Subgraphs of Cayley Graphs overp-Groups

The subject of this paper is the size of the largest component in random subgraphs of Cayley graphs, X_n, taken over a class of p -groups, G_n. G_nconsists of p -groups, G_n, with the following properties: (i)G_n / Φ(G_n)  = F_pⁿ, where Φ(G_n) is the Frattini subgroup and (ii) | G_n|  ≤ n^Kn, where K is some positive constant. We consider Cayley graphs X_n = Γ(G_n, S_n′), where S_n′ = S_n  S_n ^− 1, and S_nis a minimal G_n-generating set. By selecting G_n-elements with the independent probability λ_nwe induce random subgraphs of X_n. Our main result is, that there exists a positive constant c >  0 such that for λ_n = c ln(| S_n′ |) / | S_n′ | the largest component of random induced subgraphs of X_ncontains almost all vertices.     

有限连通图上随机扩张森林和连通子图的边负相关性

设G为有限连通图．本文研究图G的子图空间G上的三类概率测度,它们分别刻画图的随机扩张树,随机扩张森林和随机连通子图．基于G上均匀扩张树的边负相关性,我们构造G上的一族边负相关的非平凡随机扩张森林和随机连通子图．此外,我们还给出一定条件下图上均匀扩张森林的边负相关性．     

On the Number of Subgraphs of a Specified Form Embedded in a Random Graph

Let U ₁, U ₂,... be a sequence of i.i.d. random elements in R^d. For x>0, a graph G _n (x) may be formed by connecting with an edge each pair of points in that are separated by a distance no greater than x. The points of G _n (x) could represent the stations in a telecommunications network and the edge set the lines of communication that exist among them. Let be a collection of graphs on mn points having a specified form or structure, and let denote the number of subgraphs embedded in G _n (x) and contained in . It is shown that a SLLN, CLT and LIL for follow easily from the theory of U-statistics. In addition, a uniform (in x) SLLN is proved for collections that satisfy a certain monotonicity condition. Some applications are mentioned and the results of some simulations presented. The scaling constants appearing in the CLT are usually hard to obtain. These are worked out for some special cases.     

Random Subgraphs Of Finite Graphs: III. The Phase Transition For The n-Cube

We study random subgraphs of the n-cube {0,1}ⁿ, where nearest-neighbor edges are occupied with probability p. Let p_c(n) be the value of p for which the expected size of the component containing a fixed vertex attains the value λ2^n/3, where λ is a small positive constant. Let ε=n(p−p_c(n)). In two previous papers, we showed that the largest component inside a scaling window given by |ε|=Θ(2^−n/3) is of size Θ(2^2n/3), below this scaling window it is at most 2(log 2)nε⁻², and above this scaling window it is at most O(ε2ⁿ). In this paper, we prove that for the size of the largest component is at least Θ(ε2ⁿ), which is of the same order as the upper bound. The proof is based on a method that has come to be known as “sprinkling,” and relies heavily on the specific geometry of the n-cube.     

Largest Non-Unique Subgraphs

The reduction number r(G) of a graph G is the maximum integer m≤|E(G)| such that the graphs G−E, E⊆E(G),|E|≤m, are mutually non-isomorphic, i.e., each graph is unique as a subgraph of G. We prove that and show by probabilistic methods that r(G) can come close to this bound for large orders. By direct construction, we exhibit graphs with large reduction number, although somewhat smaller than the upper bound. We also discuss similarities to a parameter introduced by Erdős and Rényi capturing the degree of asymmetry of a graph, and we consider graphs with few circuits in some detail. Supported by a grant from the Danish Natural Science Research Council.     

Subgraphs and well-quasi-ordering

Let ?? be a class of graphs and let ? be the subgraph or the induced subgraph relation. We call ? an ideal (with respect to ?) if ? implies that ?. In this paper, we study the ideals that are well-quasiordered by ?. The following are our main results. If ? is the subgraph relation, we characterize the well-quasi-ordered ideals in terms of exluding subgraphs. If?is the induced subgraph relation, we present three wellquasi-ordered ideals. We also construct examples to disprove some of the possible generalizations of our results. The connections between some of our results and digraphs are considered in this paper too.     

Counting Unbranched Subgraphs

Given an arbitrary finite graph, the polynomial associates a weight zcardF to each unbranched subgraph F of length cardF. We show that all the zeros of Q have negative real part.     

Highly Symmetric Subgraphs of Hypercubes

Two questions are considered, namely (i) How many colors are needed for a coloring of the n-cube without monochromatic quadrangles or hexagons? We show that four colors suffice and thereby settle a problem of Erdös. (ii) Which vertex-transitive induced subgraphs does a hypercube have? An interesting graph has come up in this context: If we delete a Hamming code from the 7-cube, the resulting graph is 6-regular, vertex-transitive and its edges can be two-colored such that the two monochromatic subgraphs are isomorphic, cubic, edge-transitive, nonvertex-transitive graphs of girth 10.     

Subgraphs of colour-critical graphs

Some problems and results on the distribution of subgraphs in colour-critical graphs are discussed. In section 3 arbitrarily largek-critical graphs withn vertices are constructed such that, in order to reduce the chromatic number tok−2, at leastc _k n ² edges must be removed. In section 4 it is proved that a 4-critical graph withn vertices contains at mostn triangles. Further it is proved that ak-critical graph which is not a complete graph contains a (k−1)-critical graph which is not a complete graph.     

Bipartite Subgraphs and Quasi-Randomness

We say that a family of graphs is p-quasi-random, 0<p<1, if it shares typical properties of the random graph G(n,p); for a definition, see below. We denote by the class of all graphs H for which and the number of not necessarily induced labeled copies of H in G_n is at most (1+o(1))p^e(H)n^v(H) imply that is p-quasi-random. In this note, we show that all complete bipartite graphs K_a,b, a,b2, belong to for all 0<p<1.Acknowledgments We would like to thank Andrew Thomason for fruitful discussions and Yoshi Kohayakawa for organizing Extended Workshop on Combinatorics in eq5 Paulo, Ubatuba, and Rio de Janeiro, where a part of this work was done. We also thank the referees for their careful work.The first author was partially supported by NSF grant INT-0072064The second author was partially supported by NSF grants DMS-9970622, DMS-0301228 and INT-0072064Final version received: October 24, 2003     

On Subgraphs in Distance-Regular Graphs

Terwilliger [15] has given the diameter bound d (s – 1)(k – 1) + 1 for distance-regular graphs with girth 2s and valency k. We show that the only distance-regular graphs with even girth which reach this bound are the hypercubes and the doubled Odd graphs. Also we improve this bound for bipartite distance-regular graphs. Weichsel [17] conjectures that the only distance-regular subgraphs of a hypercube are the even polygons, the hypercubes and the doubled Odd graphs and proves this in the case of girth 4. We show that the only distance-regular subgraphs of a hypercube with girth 6 are the doubled Odd graphs. If the girth is equal to 8, then its valency is at most 12.     

Minimum Clique-Free Subgraphs of Kneser Graphs

Doklady Mathematics - The saturation and weak saturation numbers of Kneser graphs with clique patterns are estimated.     

Subgraphs of graphs, I

Let G and H be two simple graphs on p vertices. We give a sufficient condition, based on the minimum degree of the vertices of G and the maximum degree of the vertices of H, for H to be a subgraph of G.     

Subgraphs smaller than the girth

For graphs A, B, let () denote the number of subsets of nodes of A for which the induced subgraph is B. If G and H both have girth > k, and if () = () for every k-node tree T, then for every k-node forest F, () = (). Say the spread of a tree is the number of nodes in a longest path. If G is regular of degree d, on n nodes, with girth > k, and if F is a forest of total spread ≤k, then the value of () depends only on n and d.     

Contractible Small Subgraphs in k-connected Graphs

We prove that if G is highly connected, then either G contains a non-separating connected subgraph of order three or else G contains a small obstruction for the above conclusion. More precisely, we prove that if G is k-connected (with k ≥ 2), then G contains either a connected subgraph of order three whose contraction results in a k-connected graph (i.e., keeps the connectivity) or a subdivision of ${K_4^-}$ whose order is at most 6.     

γ-可图序列与γ-完全子图

一个r-图是一个无环的无向图,其中任何两个顶点之间至多被r条边连接.一个m+1个顶点的r-完全图,记为K_(m+1)^((r)),是一个m+1个顶点的r-图,其中任何两个顶点之间恰好被r条边连接.一个非增的非负整数序列π=(d_1,d_2,…,d_n)称为是r-可图的如果它是某个n个顶点的r-图的度序列.一个r-可图序列π称为是蕴含(强迫)K_(m+1)((r)),是一个m+1个顶点的r-图,其中任何两个顶点之间恰好被r条边连接.一个非增的非负整数序列π=(d_1,d_2,…,d_n)称为是r-可图的如果它是某个n个顶点的r-图的度序列.一个r-可图序列π称为是蕴含(强迫)K_(m+1)^((r))可图的如果π有一个实现包含K_(m+1)((r))可图的如果π有一个实现包含K_(m+1)^((r))作为子图(π的每一个实现包含K_(m+1)((r))作为子图(π的每一个实现包含K_(m+1)^((r))作为子图).设σ(K_(m+1)((r))作为子图).设σ(K_(m+1)^((r)),n)(τ(K_(m+1)((r)),n)(τ(K_(m+1)^((r)),n))表示最小的偶整数t,使得每一个r-可图序列π=(d_1,d_2,…,d_n)具有∑_(i=1)((r)),n))表示最小的偶整数t,使得每一个r-可图序列π=(d_1,d_2,…,d_n)具有∑_(i=1)ⁿ d_i≥t是蕴含(强迫)K_(m+1)n d_i≥t是蕴含(强迫)K_(m+1)^((r))-可图的.易见,σ(K_(m+1)((r))-可图的.易见,σ(K_(m+1)^((r)),n)是Erds等人的一个猜想从1-图到r-图的扩充且τ(K_(m+1)((r)),n)是Erds等人的一个猜想从1-图到r-图的扩充且τ(K_(m+1)^((r)),n)是经典Turan定理从1-图到r-图的扩充.本文给出了蕴含K_(m+1)((r)),n)是经典Turan定理从1-图到r-图的扩充.本文给出了蕴含K_(m+1)^((r))的r-可图序列的两个简单充分条件.此两个条件包含了Yin和Li在[Discrete Math.,2005,301:218-227]中的两个主要结果和当n≥max{m((r))的r-可图序列的两个简单充分条件.此两个条件包含了Yin和Li在[Discrete Math.,2005,301:218-227]中的两个主要结果和当n≥max{m²+3m+1-[(m2+3m+1-[(m²+m)/r],2m+1+[m/r]]}时,σ(K_(m+1)2+m)/r],2m+1+[m/r]]}时,σ(K_(m+1)^((r)),n)之值.此外,我们还确定了当n≥m+1时,τ(K_(m+1)((r)),n)之值.此外,我们还确定了当n≥m+1时,τ(K_(m+1)^((r)),n)之值.     

Ramsey-type Theorems with Forbidden Subgraphs

Dedicated to the memory of Paul Erdős   A graph is called H-free if it contains no induced copy of H. We discuss the following question raised by Erdős and Hajnal. Is it true that for every graph H, there exists an such that any H-free graph with n vertices contains either a complete or an empty subgraph of size at least ? We answer this question in the affirmative for a special class of graphs, and give an equivalent reformulation for tournaments. In order to prove the equivalence, we establish several Ramsey type results for tournaments. Received August 22, 1999 RID="*" ID="*" Supported by a USA Israeli BSF grant, by a grant from the Israel Science Foundation and by the Hermann Minkowski Minerva Center for Geometry at Tel Aviv University. RID="†" ID="†" Supported by NSF grant CR-9732101, PSC-CUNY Research Award 663472, and OTKA-T-020914. RID="‡" ID="‡" Supported by TKI grant Stochastics@TUB, and OTKA-T-026203.     

Forbidden Induced Subgraphs for Toughness

Let be a family of connected graphs. A graph G is said to be ‐free if G is H‐free for every graph H in . We study the relation between forbidden subgraphs in a connected graph G and the resulting toughness of G. In particular, we consider the problem of characterizing the graph families such that every large enough connected ‐free graph is t‐tough. In this article, we solve this problem for every real positive number t. © 2012 Wiley Periodicals, Inc. J. Graph Theory 73: 191–202, 2013