ONLINE VERTEX COLORINGS OF RANDOM GRAPHS WITHOUT MONOCHROMATIC SUBGRAPHS

Size: px
Start display at page:

Download "ONLINE VERTEX COLORINGS OF RANDOM GRAPHS WITHOUT MONOCHROMATIC SUBGRAPHS"

Transcription

1 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS WITHOUT MONOCHROMATIC SUBGRAPHS MARTIN MARCINISZYN AND RETO SPÖHEL Abstract. Consider the following generalized notion of graph colorings: a vertex coloring of graph G is valid w.r.t. some fixed nonempty graph F if no color class induces a copy of F in G, i.e., there is no monochromatic copy of F in G. We propose and analyze an algorithm for computing valid colorings of a random graph G n,p on n vertices with edge probability p in an online fashion. For a large family of graphs F including cliques and cycles of arbitrary size, the proposed algorithm is optimal in the following sense: for any integer r 1, there is a constant β = β(f, r) such that the algorithm a.a.s. (asymptotically almost surely) computes a valid r-coloring of G n,p w.r.t. F online if p n β, and any online algorithm will a.a.s. fail to do so if p n β. That is, we observe a threshold phenomenon determined by the function n β. 1. Introduction The chromatic number χ(g) of a graph G is the minimum number of colors r such that there exists an r-vertex-coloring of G in which no edge has endpoints of the same color. Computing the chromatic number of a graph is a well-known NP-complete problem [8]. Moreover, very deep results [5, 6, 13] show that we cannot even efficiently approximate the value of χ(g) of graphs G on n vertices within a factor of n 1 ε for any constant ε > 0 unless NP ZP P. (ZP P is the class of languages decidable by a random expected polynomial-time algorithm that makes no errors.) Many bounds on the chromatic number can be found in the literature. Clearly, the size of the largest clique ω(g) in graph G is a lower bound on χ(g). On the other hand, (G) + 1 is an upper bound on the chromatic number, where (G) denotes the maximum degree in G as usual. This bound can be obtained from a simple first fit algorithm. Suppose G is a graph on the vertex set {v 1,..., v n }, and r := (G) + 1 colors are identified by numbers 1,..., r. First fit considers the vertices of G in the order v 1,..., v n. To each vertex v i, 1 i n, it assigns the smallest color available that does not create a monochromatic edge, i.e., the smallest color index k, 1 k r, such that no vertex v j adjacent to v i with j < i was assigned to k before. This obviously produces a valid r-coloring of G. Employing a more intelligent ordering of the vertices, the same algorithm yields the theorem of Brooks [3]. This classical result states that for every graph G, we have χ(g) (G) unless G contains a clique on (G) + 1 vertices or an odd cycle if (G) = 2. We are interested in online-colorings of random graphs. We consider the following graph process starting with the empty graph G 0. In each step i > 0, a new vertex v i appears and is randomly connected to the vertices of G i 1 by independently including each of the edges {v i, v j }, 1 j < i with probability p into a new graph G i. After n steps this process results in a random graph G = G n,p, in which each edge is present with probability p independently of all other edges. An online coloring strategy must assign a color to v i in each step i immediately, thereby extending the coloring of G i 1 to G i in a proper way. It is well-known [2, 9, 11] that in random graphs with p n 1, first fit asymptotically almost surely (a.a.s., i.e., with probability tending to 1 as n goes to infinity) needs about 2χ(G n,p ) colors. Institute of Theoretical Computer Science, ETH Zurich, 8092 Zurich, Switzerland {mmarcini,rspoehel}@inf.ethz.ch. The second author was supported by SNF grant

2 2 MARTIN MARCINISZYN AND RETO SPÖHEL In sparser random graphs, first fit performs not that well. For p = c/n, χ(g n,p ) is a.a.s. bounded by a constant only depending on c. However, Pittel and Weishaar [16] showed that first fit a.a.s. uses about log 2 log n colors in random graphs of this density. We view the question from the opposite perspective: supposing that only a constant number of colors r is available, we ask for which values of p an online coloring strategy exists that a.a.s. produces a valid coloring of G n,p. In this paper, we primarily consider a more general notion of valid colorings than described above. For the classical case, our results yield that first fit a.a.s. colors G n,p with r colors as long as p n 1/(1 2 r), and that no online algorithm can do substantially better. The maximal degree in G n,p with p slightly below this bound is roughly 2 r, so first fit a.a.s. performs exponentially better than what is guaranteed by Brooks theorem. On the other hand, G = G n,p is a.a.s. a forest and has chromatic number χ(g) = 2 for p n 1. In this paper, we consider the following widely used generalized notion of vertex colorings of graphs. Rather than requiring that none of the color classes induces an edge in G, we call an r- coloring valid w.r.t. a fixed nonempty graph F if no color class induces a copy of F in G, i.e., there is no monochromatic copy of F in G. Suppose r 1 is a fixed integer. We propose and analyze an algorithm for computing valid r-colorings of random graphs online. For notational reasons that will become clear later, we consider the r available colors in the reverse order r,..., 1. The simple greedy F -avoidance strategy of selecting the largest color index that allows for a valid extension of the coloring turns out to be not optimal in general. For example, consider the graph F consisting of a triangle with one edge attached to it. Here our results yield that greedily avoiding triangles and forgetting about the additional edge is better than greedily avoiding the entire graph. Therefore, we consider the following refinement. Let H 1,..., H r be a sequence of fixed nonempty subgraphs H k F, 1 k r. Then the greedy H 1,..., H r -avoidance strategy chooses, in each step i, the maximal color index k such that v i is not contained in a copy of H k induced by the color class k in G i. In other words, it avoids monochromatic copies of H k in color k, 1 k r. (The inclusion of H 1 may seem artificial but turns out to be convenient.) Clearly, if we succeed in assigning a color to each vertex of G in this way, we obtain a valid r-coloring w.r.t. the graph F. We show how to choose the subgraphs H 1,..., H r such that, for a large family of graphs F, the resulting greedy H 1,..., H r -avoidance strategy is optimal in the following sense: there is a constant β = β(f, r) such that the greedy H 1,..., H r -avoidance strategy a.a.s. computes a valid r-coloring w.r.t. F of G n,p if p n β, and any online strategy will a.a.s. fail to do so if p n β. That is, we observe a threshold phenomenon determined by the function n β. More precisely, the value of β is reciprocal to the following recursively defined density measure for graphs, where we denote E(H) by e H and V (H) by v H : m 0 1(F ) := 0 and m k 1(F ) := max H F e H + m k 1 1 (F ) v H k 1. (1) For any given graph F, we define the following infinite sequence of graphs accordingly: Our main result reads as follows. H k = H k (F ) := argmax H F e H + m k 1 1 (F ) v H k 1. (2) Theorem 1. Let F be a nonempty graph and r 1 an integer. Let H 1,..., H r F be chosen according to (2). Then the greedy H 1,..., H r -avoidance strategy a.a.s. produces an r-coloring of G n,p without any monochromatic copy of F online provided we have p = p(n) n 1/mr 1 (F ).

3 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS 3 Moreover, assuming there exists an induced subgraph F F on v F 1 vertices that satisfies this strategy is optimal. e H max H F v H 1 m2 1(F ), (3) We refer to the greedy H 1,..., H r -avoidance strategy with H 1,..., H r as in (2) as the enhanced first fit algorithm. Assumption (3) is inherent to our method for proving the optimality of this strategy. It holds for a large family of graphs including cliques and cycles of arbitrary size. In these cases, we also obtain a closed form for the function m r 1 (F ). As usual, let K l and C l denote a clique and a cycle on l vertices respectively. As those graphs have a highly regular structure, we have H k = F for all k 1, i.e., the simple first fit algorithm is optimal. Corollary 2 (Cliques). For all l 2 and r 1, the greedy K l -avoidance strategy a.a.s. produces an r-coloring of G n,p without any monochromatic clique of size l online provided we have and this strategy is optimal. p = p(n) n 2 l(1 l r ), Corollary 3 (Cycles). For all l 2 and r 1, the greedy C l -avoidance strategy a.a.s. produces an r-coloring of G n,p without any monochromatic cycle of length l online provided we have and this strategy is optimal. p = p(n) n l 1 l(1 l r ), 1.1. Bounds on the threshold function. Two classical results on random graphs yield obvious lower and upper bounds on the threshold for online-colorability with a fixed number of colors r. Clearly, if after the last step of the online process G = G n,p contains no copy of F, extending the coloring of any intermediate graph G i from i to i + 1 vertices is trivial. The well-known theorem of Bollobàs [1], which is a generalization of a result of Erdős and Rényi [4] to arbitrary graphs F, states a threshold for this event. Theorem 4. Let F be a nonempty graph. Then we have lim P[F G n,p] = n { 0 if p n 1/m(F ) 1 if p n 1/m(F ), where e H m(f ) := max. H F v H Therefore, for p n 1/m(F ), there is a.a.s. no copy of F in G n,p, and any coloring algorithm will succeed. In fact, we have m 1 1 (F ) = m(f ), and setting r = 1 in Theorem 1 yields the statement of Theorem 4. It is easy to see that for all r 2, m r 1 (F ) is strictly greater than m(f ) and thus yields a higher threshold. An upper bound on the threshold is obtained by studying offline colorings of random graphs. Obviously, the existence of a valid r-coloring of G = G n,p is necessary for any strategy to succeed. A result of Luczak, Ruciński, and Voigt [12] states that, beyond a certain threshold for the edge probability p, G a.a.s. admits no r-coloring without a monochromatic copy of F. Theorem 5. Let r 2 and F be a nonempty graph that is not a matching if r = 2. Moreover, let G (F ) v r denote the property that every r-coloring of G contains a monochromatic copy of F.

4 4 MARTIN MARCINISZYN AND RETO SPÖHEL Then there exist positive constants c = c(f, r) and C = C(F, r) such that { lim P[G n,p (F ) v 0 if p cn 1/m 1(F ) n r] = 1 if p Cn 1/m, 1(F ) where m 1 (F ) := max H F e H v H 1. Hence, there is no hope of finding an r-coloring avoiding a monochromatic copy of F whenever p n 1/m 1(F ). Note that the function m 1 (F ) determining the order of magnitude of the threshold does not depend on the number of available colors r in the offline case. In the online case, however, the density function m r 1 (F ) is monotone increasing with the number of available colors r. It is easily deduced from the formulas in Corollaries 2 and 3 that, in the case of cliques and cycles, the online threshold converges to the offline threshold as r tends to infinity. One can show that this convergence holds in general, i.e., that all graphs F satisfy lim r mr 1(F ) = m 1 (F ). It follows that for any constant ε > 0, the enhanced first fit algorithm a.a.s. produces a valid r-vertex-coloring of G n,p with p n 1/m 1(F ) ε if r is sufficiently large Related Work. The problem of finding thresholds for online-colorability with a fixed number of colors was first considered for edge-colorings of random graphs. Investigating algorithmic Ramsey properties of triangles, Friedgut et al. [7] introduced and studied the online edge coloring game. In this game we start with an empty graphs on n vertices and new edges arrive in a random order. A single player must assign one of r available colors to each new edge immediately such that no monochromatic copy of a given graph F is created. The game ends as soon as the player is forced to close such a monochromatic copy. The authors of [7] showed that there exists a threshold for the duration of the game in the case when F is a triangle and r = 2. They proved that the player can a.a.s. avoid monochromatic triangles for any N n 4/3 random edges by playing greedily. If, however, N n 4/3 random edges are inserted, the player a.a.s. loses the game no matter which strategy is chosen. This result was generalized in [14] and [15] to a large family of graphs F including cliques and cycles of arbitrary size. It was shown that for those graphs F, the threshold for the duration of the F -avoidance edge coloring game is given by where N 0 (n) := n 2 1/m 2(F ), m 2 (F ) := max H F e H v H 2 + 1/m(F ). It seems plausible that m 2 (F ) determines a threshold for all graphs F other than forests, and that the theory generalizes to the game with more than two colors as in the vertex-coloring case. However, for r 3, there are only non-trivial lower bounds on the duration of the F -avoidance edge coloring game. 2. Preliminaries and notation The expression f g means that the functions f and g differ at most by a multiplicative constant. All graphs are simple and undirected. The number of vertices of a graph G is denoted by v G or v(g), and similarly the number of edges by e G or e(g).

5 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS Density measures for graphs. A density measure is a function mapping graphs to nonnegative reals. For simplicity, we define that all density measures evaluate to zero on empty graphs. The most well-known density measure is d(h) := e H v H for any non-empty graph H. This is exactly the average degree of H divided by 2. The density measure d 1 (H) := e H v H 1 is also well-known and motivated by Theorem 5. For any given density measure d i, we define the maximum density m i (H) := max d i(j) J H (cf. Theorems 4 and 5). We say that H is balanced with respect to d i if m i (H) = d i (H). H is balanced in the ordinary sense or simply balanced if it is balanced w.r.t. the density measure d. We call H 1-balanced if it is balanced w.r.t. the density measure d 1. It is well-known that every graph H satisfies m(h) m 1 (H), and that every 1-balanced graph is also balanced. Reflecting the maximum density measure m r 1 (H) as defined in (1), for any r 1 and nonempty graphs H and J, let d r 1(H, J) := e J + m r 1 1 (H) v J. We abbreviate d r 1(H, H) by d r 1(H). We say that H is balanced w.r.t. d r 1 if m r 1 (H) = dr 1(H). By definition, we have m 1 1 (H) = m(h) and d1 1(H) = d(h). We use the shorthand notation m 1 for m 2 1 and d 1 for d 2 1 respectively. The maximum density m r 1 was apparently not studied before in the literature. The next lemma shows that it is sandwiched between m and m 1. The proof involves only elementary calculations and is omitted in this extended abstract. Lemma 6. (i) Every nonempty graph F satisfies (ii) Every graph F satisfies m(f ) = m 1 1(F ) < m 2 1(F ) < < m r 1(F ) < < m 1 (F ). lim r mr 1(F ) = m 1 (F ). (iii) Every 1-balanced graph is balanced with respect to d r 1 for all r 1. (iv) For r 2, every graph that is balanced with respect to d r 1 is balanced with respect to d r 1 1. It is easily verified that the inductive definition of m r 1 (F ) can be written in the following explicit form, which yields the threshold formulas in Corollaries 2 and 3. The proof proceeds by induction on r. Lemma 7. For all nonempty graphs F and r 1, we have If F is 1-balanced, we have m r 1(F ) = max H 1,...,H r F m r 1(F ) = r i=1 e(h i) i 1 j=1 v(h j) r i=1 v(h i) e F v F 1 ( ) 1 v r F.. (4)

6 6 MARTIN MARCINISZYN AND RETO SPÖHEL 2.2. Janson s inequality. Janson s inequality [10] is a very useful tool in probabilistic combinatorics. In many cases, it yields an exponential bound on lower tails where the second moment method only gives a bound of o(1). Here we formulate a version tailored to random graphs. Theorem 8. Consider a family F of graphs on the vertex set [n]. For each H F, let X H denote the indicator random variable for the event H G n,p, and for each pair H 1, H 2 F, H 1 H 2, write H 1 H 2 if H 1 and H 2 are not edge-disjoint. Let X = H F X H, µ = E[X] = H F p e(h), = H 1,H 2 F H 1 H 2 E[X H1 X H2 ] = H 1,H 2 F H 1 H 2 e(h p 1 )+e(h 2 ) e(h 1 H 2 ). Then, for all 0 δ 1, we have P[X (1 δ)µ] e δ 2 µ 2 2(µ+ ). 3. Analysis of the enhanced first fit algorithm In this section we analyze the enhanced first fit algorithm as proposed in the introduction and show that it a.a.s. produces valid r-colorings of G n,p w.r.t. F up to the edge probability p 0 = p 0 (F, r, n) := n 1/mr 1 (F ). We present the proofs from a game perspective as in Section 1.2. A single player called Painter is given the vertices of a random graph one by one in the manner described in the introduction, and has to color them immediately. Her aim is to color the entire graph without creating a monochromatic copy of some fixed graph F. Consider the game with two colors, say red (2) and blue (1), and suppose Painter plays the greedy H 1, H 2 -avoidance strategy with H 1, H 2 F fixed. That is, she uses red in every step unless this creates a red copy of H 2. Note that choosing H 1 F implies that the game comes to a premature end as soon as a blue copy of H 1 is created. As will become evident soon, this subtle technicality makes no substantial difference provided H 1 is suitably chosen, but greatly improves readability. Thus, when Painter loses the game, the random graph contains an entirely blue copy of H 1 with the following property: recoloring any vertex of this copy of H 1 to red creates an entirely red copy of H 2. We say that every vertex of this copy of H 1 closes a red copy of H 2. This motivates the definition of the following family of graphs. Let F(H 1, H 2 ) denote the class of all graphs that have an inner (blue) copy of H 1, each vertex of which closes an outer (red) copy of H 2. Here the colors should only provide the intuitive connection to the greedy strategy, the members of the family F(H 1, H 2 ) are not associated with a coloring. Two graphs from the family F(K 4, K 4 ) are depicted in Figure 1. We say that the inner copy of H 1 is formed by inner vertices and edges, and refer to the surrounding elements as outer vertices and edges respectively. Observe that in this terminology every outer copy contains exactly one inner vertex and no inner edges. We formally define this family of graphs as follows.

7 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS 7 Figure 1. Two graphs from the family F(K 4, K 4 ). The graph on the left hand side is the unique member with vertex-disjoint outer copies. In both graphs, the inner vertices and edges are shaded. Definition 9. For all graphs H 1 = (V, E) and H 2, let { F(H 1, H 2 ) := F = (V. U, E. D) : F is a minimal graph such that for all v V there are sets U(v) U and D(v) D with ( v. ) } U(v), D(v) = H2 We take F as a minimal element with respect to subgraph inclusion, i.e., F does not have a subgraph which satisfies the same properties. This ensures in particular that F(H 1, H 2 ) is finite. In the game with three colors, say, with colors yellow (3), red (2) and blue (1), a greedy Painter first avoids some subgraph H 3 in yellow, H 2 in red, and eventually H 1 in blue. By the same argument as in the case of two colors, when Painter loses the game, the random graph contains a red-blue copy of a member from the family F(H 1, H 2 ), every vertex of which closes an entirely yellow copy of H 3. We denote the class of all such graphs by F(H 1, H 2, H 3 ). This motivates the following inductive definition for general r. For notational reasons, we include the game with one color into the formalism. Definition 10. For any graph H 1, let F(H 1 ) := {H 1 }. For r 2 and arbitrary graphs H 1,..., H r, let F(H 1,..., H r ) := { F F(F, H r ) : F F(H 1,..., H r 1 ) }. With these definitions at hand, we can state the following simple observation. Lemma 11. Let F be a nonempty graph and r 1 an integer. Let H 1,..., H r F be arbitrary nonempty subgraphs. Then Painter a.a.s. wins the online vertex coloring game w.r.t. F with r colors if playing the greedy H 1,..., H r -avoidance strategy as long as p n 1/g(H 1,H 2,...,H r), where g(h 1, H 2,..., H r ) := min { m(f ) : F F(H 1, H 2,..., H r ) }. Proof. As argued before, a member of the family F(H 1,..., H r ) must be contained in G n,p if Painter loses greedily avoiding H 1,..., H r. By Theorem 4 and the definition of g, such a graph does a.a.s. not exist in G n,p with p n 1/g(H 1,...,H r). Now, the first part of Theorem 1 follows from Lemma 11 and the next lemma. Lemma 12. Let F be a nonempty graph and r 1 an integer. If the sequence of subgraphs H 1,..., H r is chosen according to (2), all graphs F F(H 1,..., H r ) satisfy m(f ) m r 1(F )..

8 8 MARTIN MARCINISZYN AND RETO SPÖHEL Lemmas 11 and 12 imply together that Painter can a.a.s. successfully color the entire graph G n,p without any monochromatic copy of F online as long as p n 1/mr 1 (F ). It remains to prove Lemma 12. In this extended abstract, we only outline the underlying ideas of this proof. For the sake of simplicity, consider the case r = 2. Among the members of the family F(H 1, H 2 ), there are some distinguished ones that have vertex-disjoint outer copies of H 2. The left hand side of Figure 1 shows an example for the case H 1 = H 2 = K 4. Consider any such member F 2 F(H 1, H 2 ) with vertex-disjoint outer copies. Clearly, we have due to the choice of H 1 and H 2. graphs F F(H 1, H 2 ), we have e(f 2 ) v(f 2 ) = v H 1 e H2 + e H1 v H1 v H2 = e H 2 + e H1 /v H1 v H2 = m 1 (F ) Hence, Lemma 12 is equivalent to the assertion that for all m(f ) e(f 2 ) v(f2 ). Thus, the statement of Lemma 12 is essentially that members of the family F(H 1, H 2 ) that contain overlapping substructures are at least as dense as the nice members F2. A formal proof is not hard to obtain, but slightly technical. It is omitted in this extended abstract. For r 3, the lemma follows by induction. Note that the right hand side of (4) is exactly the density of a nice graph Fr F(H 1,..., H r ) in which all outer copies are vertex-disjoint. 4. Optimality of the enhanced first fit algorithm In this section, we prove that the enhanced first fit algorithm is optimal in the sense that any online algorithm a.a.s. fails to produce a valid r-coloring of G n,p w.r.t. F if p n 1/mr 1 (F ). We employ a multi-round exposure approach, in each round using the following counting version of Theorem 5. The proof of Theorem 13 proceeds along the same lines as the upper bound proof of Theorem 5, utilizing Janson s inequality. It is omitted due to space restrictions. Theorem 13. Let r 1 and F be a non-empty graph. Then there exist constants C = C(F, r) > 0 and a = a(f, r) > 0 such that for all p(n) Cn 1/m 1(F ), a.a.s. in every r-vertex-coloring of G n,p, one color class induces at least an v F p e F monochromatic copies of F. The assumption that there exists an induced subgraph F F on v F 1 vertices satisfying m 1 (F ) m 1 (F ) (cf. (3)) guarantees that we can apply Theorem 13 to F. Rather than generating the random graph G n,p online as stated in the introduction, it is more convenient here to generate the entire graph in advance. The random experiment is conducted for all edges before the game starts. In each move i, one new vertex v i is revealed to Painter together with all edges connecting v i to the vertices v j, 1 j < i. Edges that will appear in the future are hidden from Painter s view. Assume that p n 1/mr 1 (F ) is given. We prove by induction on r that Painter a.a.s. loses the game regardless of the strategy she employs. The case r = 1 follows directly from Theorem 4. Suppose we have r 2 and the statement holds for r 1. We partition the vertex set of G = G n,p into two sets V 1 := {v 1,..., v n/2 } and V 2 = {v n/2 +1,..., v n } and relax the game to a semi-online game, revealing the subgraph G n/2 = G[V 1 ] to Painter all at once. She may color the vertices of V 1 offline. In the second round, the vertices of V 2 are revealed one by one, and Painter has to color them online as before.

9 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS 9 Suppose that Painter s coloring of the vertex set V 1 is fixed and consider the set of edges E G (V 1, V 2 ) generated between V 1 and V 2, but hidden from Painter s view. For each color k {1,..., r}, this edge set defines a vertex set Base(k) V 2 consisting of all vertices in V 2 that complete a copy F G[V 1 ] in color k to a copy of F. Obviously, Painter may not assign color k to any vertex in Base(k) in the second round. The next claim asserts that there exists some color class k 0 {1,..., r} such that Base(k 0 ) is sufficiently large so that we can apply the induction hypothesis. Claim 14. After the first round was played, there a.a.s. exists a color k 0 {1,..., r} such that we have Base(k 0 ) = Ω (n mr 1 1 (F )/m r 1 (F )). Claim 14 implies that in the second round, Painter must a.a.s. color a binomial random graph G en,ep on ñ n mr 1 1 (F )/m r 1 (F ) vertices with edge probability p = p n 1/mr 1 (F ) ñ 1/mr 1 1 (F ) with just r 1 colors left in an online fashion. Applying the induction hypothesis yields that she a.a.s. fails no matter which strategy she employs. Hence, it remains to prove Claim 14. Proof of Claim 14. In order to simplify notation, let β := m r 1 1 (F )/m r 1 (F ). Note that we have β < 1 due to Lemma 6. For any graph J F with v J 1 and p p 0 := n 1/mr 1 (F ), we have n v J p e J n v J e J /m r 1 (F ) n v J = n v J e J +v J m r 1 1 (F ) e J v J e J +m r 1 1 (F ) = n e J v J e J +m r 1 1 (F ) v J e J +m r 1 1 (F ) mr 1 1 (F ) n β, (5) with equality if p = p 0 and d r 1(F, J) = m r 1 (F ). By Theorem 13 and owing to m 1 (F ) m 1 (F ) m r 1 (F ) (cf. (3)), Painter s offline-coloring of the vertices in V 1 a.a.s. creates Ω ( n v(f ) p e(f ) ) monochromatic copies of F. By the first moment method, there are a.a.s. at most O ( n v(f ) p e(f ) ) copies of F in total, and therefore there are a.a.s. Θ ( n v(f ) p e(f ) ) monochromatic copies of F in G[V 1 ]. W.l.o.g. we assume that those copies have color r. Suppose F 2 is a graph consisting of the union of two copies of F intersecting on some nonempty graph J F. Then there are a.a.s. at most O ( n 2v(F ) v(j ) p 2e(F ) e(j ) ) (6) copies of F 2 in G[V 1 ] by the first moment method. Since there are only constantly many nonisomorphic graphs F 2 with this structure, the bound in (6) a.a.s. hold for all of them in G[V 1 ]. Next, we estimate the probability that a fixed vertex v V 2 is contained in Base(r). Let u denote the missing vertex in F, i.e., {u} = V (F ) \ V (F ). For any subgraph J F, let Ĵ F denote the graph in which all edges leading from J to u are added, and let deg bj (u) = e(ĵ) e(j ) denote the degree of u in Ĵ. For any vertex v V 2 and any monochromatic copy Fi of F in G[V 1 ], let F v,i be the family of all minimal subsets of V 1 V 2 that connect Fi and v such that those form a copy of F. Note that F v,i has constant size depending only on the structure of F and F. If F is a clique, e.g., F v,i contains at most one member. Let F v := F i V 1 F v,i

10 10 MARTIN MARCINISZYN AND RETO SPÖHEL be a multi-set allowing for duplicate entries since the same set of edges may complement distinct monochromatic copies of F in G[V 1 ] to F. For each member H F v, let X H denote the indicator random variable for the event H G n,p. Let X v := H F v X H. Clearly, the vertex v is contained in Base(r) if X v 1. We apply Theorem 8 to the family F v in order to obtain a lower bound on the probability of this event. Simple calculations yield that we have ( µ = E[X v ] = Θ n v(f ) ) p e(f ) p deg F (u) = Θ ( n v F 1 p e ) ( (5) F =Ω n β 1). Moreover, a.a.s. we have = H 1,H 2 F v E[X X ] H1 H2 H 1 H 2 Θ(1)p 2 deg F (u) deg J b(u) J F F e(j i,f j ) 1 Fi F j =J (6) = J F e(j ) 1 O ( n 2v(F ) v(j ) p 2e(F ) e(j ) ) p 2 deg F (u) deg b J (u) O ( n 2v F v J 1 p 2e ) F e J J F v J 1 = µ 2 J F v J 1 O ( n 1 v J p e ) ( (5) J =O µ 2 n 1 β). Therefore, Theorem 8 yields that { µ 2 } P[v / Base(r)] = P [X v = 0] exp 2(µ + ) { ( = exp Ω n β 1)} ( = 1 Ω n β 1). Since for two distinct vertices u, v V 2, the random variables X u and X v are independent, standard Chernoff bounds imply that a.a.s. we have Base(r) = Ω ( P [X v 1] = Ω n β). v V2 Note that this proof exhausts our approach for proving the optimality of the enhanced first fit strategy even for the game with just two colors: if (3) is violated, Painter can a.a.s. color all vertices of V 1 without creating a single monochromatic copy of F, and Base(1), Base(2) V 2 are both empty. Thus, there is little hope of proving general optimality of our strategy by considering the relaxation to a multi-round offline game as proposed in this paper. References [1] B. Bollobás. Threshold functions for small subgraphs. Math. Proc. Cambridge Philos. Soc., 90(2): , [2] B. Bollobás. The chromatic number of random graphs. Combinatorica, 8(1):49 55, [3] R. L. Brooks. On colouring the nodes of a network. Proc. Cambridge Philos. Soc., 37: , 1941.

11 ONLINE VERTEX COLORINGS OF RANDOM GRAPHS 11 [4] P. Erdős and A. Rényi. On the evolution of random graphs. Magyar Tud. Akad. Mat. Kutató Int. Közl., 5:17 61, [5] U. Feige and J. Kilian. Zero knowledge and the chromatic number. J. Comput. System Sci., 57(2): , Complexity 96 The Eleventh Annual IEEE Conference on Computational Complexity (Philadelphia, PA). [6] M. Fürer. Improved hardness results for approximating the chromatic number. In Proceedings of the Thirty-Sixth Annual Symposium on Foundations of Computer Science, pages IEEE, [7] E. Friedgut, Y. Kohayakawa, V. Rödl, A. Ruciński, and P. Tetali. Ramsey games against a one-armed bandit. Combin. Probab. Comput., 12(5-6): , Special issue on Ramsey theory. [8] M. R. Garey and D. S. Johnson. Computers and intractability. W. H. Freeman and Co., San Francisco, Calif., A guide to the theory of NP-completeness, A Series of Books in the Mathematical Sciences. [9] G. R. Grimmett and C. J. H. McDiarmid. On colouring random graphs. Math. Proc. Cambridge Philos. Soc., 77: , [10] S. Janson. Poisson approximation for large deviations. Random Structures Algorithms, 1(2): , [11] T. Luczak. The chromatic number of random graphs. Combinatorica, 11(1):45 54, [12] T. Luczak, A. Ruciński, and B. Voigt. Ramsey properties of random graphs. J. Combin. Theory Ser. B, 56(1):55 68, [13] C. Lund and M. Yannakakis. On the hardness of approximating minimization problems. J. Assoc. Comput. Mach., 41(5): , [14] M. Marciniszyn, R. Spöhel, and A. Steger. The online clique avoidance game on random graphs. In Approximation, randomization and combinatorial optimization, volume 3624 of Lecture Notes in Comput. Sci., pages Springer, Berlin, [15] M. Marciniszyn, R. Spöhel, and A. Steger. Online Ramsey games in random graphs. Submitted, [16] B. Pittel and R. S. Weishaar. On-line coloring of sparse random graphs and random trees. J. Algorithms, 23(1): , 1997.

On-line Ramsey numbers

On-line Ramsey numbers On-line Ramsey numbers David Conlon Abstract Consider the following game between two players, Builder and Painter Builder draws edges one at a time and Painter colours them, in either red or blue, as each

More information

Mean Ramsey-Turán numbers

Mean Ramsey-Turán numbers Mean Ramsey-Turán numbers Raphael Yuster Department of Mathematics University of Haifa at Oranim Tivon 36006, Israel Abstract A ρ-mean coloring of a graph is a coloring of the edges such that the average

More information

A Turán Type Problem Concerning the Powers of the Degrees of a Graph

A Turán Type Problem Concerning the Powers of the Degrees of a Graph A Turán Type Problem Concerning the Powers of the Degrees of a Graph Yair Caro and Raphael Yuster Department of Mathematics University of Haifa-ORANIM, Tivon 36006, Israel. AMS Subject Classification:

More information

Best Monotone Degree Bounds for Various Graph Parameters

Best Monotone Degree Bounds for Various Graph Parameters Best Monotone Degree Bounds for Various Graph Parameters D. Bauer Department of Mathematical Sciences Stevens Institute of Technology Hoboken, NJ 07030 S. L. Hakimi Department of Electrical and Computer

More information

Large induced subgraphs with all degrees odd

Large induced subgraphs with all degrees odd Large induced subgraphs with all degrees odd A.D. Scott Department of Pure Mathematics and Mathematical Statistics, University of Cambridge, England Abstract: We prove that every connected graph of order

More information

Exponential time algorithms for graph coloring

Exponential time algorithms for graph coloring Exponential time algorithms for graph coloring Uriel Feige Lecture notes, March 14, 2011 1 Introduction Let [n] denote the set {1,..., k}. A k-labeling of vertices of a graph G(V, E) is a function V [k].

More information

A NOTE ON OFF-DIAGONAL SMALL ON-LINE RAMSEY NUMBERS FOR PATHS

A NOTE ON OFF-DIAGONAL SMALL ON-LINE RAMSEY NUMBERS FOR PATHS A NOTE ON OFF-DIAGONAL SMALL ON-LINE RAMSEY NUMBERS FOR PATHS PAWE L PRA LAT Abstract. In this note we consider the on-line Ramsey numbers R(P n, P m ) for paths. Using a high performance computing clusters,

More information

ON INDUCED SUBGRAPHS WITH ALL DEGREES ODD. 1. Introduction

ON INDUCED SUBGRAPHS WITH ALL DEGREES ODD. 1. Introduction ON INDUCED SUBGRAPHS WITH ALL DEGREES ODD A.D. SCOTT Abstract. Gallai proved that the vertex set of any graph can be partitioned into two sets, each inducing a subgraph with all degrees even. We prove

More information

On an anti-ramsey type result

On an anti-ramsey type result On an anti-ramsey type result Noga Alon, Hanno Lefmann and Vojtĕch Rödl Abstract We consider anti-ramsey type results. For a given coloring of the k-element subsets of an n-element set X, where two k-element

More information

Graphs without proper subgraphs of minimum degree 3 and short cycles

Graphs without proper subgraphs of minimum degree 3 and short cycles Graphs without proper subgraphs of minimum degree 3 and short cycles Lothar Narins, Alexey Pokrovskiy, Tibor Szabó Department of Mathematics, Freie Universität, Berlin, Germany. August 22, 2014 Abstract

More information

SEQUENCES OF MAXIMAL DEGREE VERTICES IN GRAPHS. Nickolay Khadzhiivanov, Nedyalko Nenov

SEQUENCES OF MAXIMAL DEGREE VERTICES IN GRAPHS. Nickolay Khadzhiivanov, Nedyalko Nenov Serdica Math. J. 30 (2004), 95 102 SEQUENCES OF MAXIMAL DEGREE VERTICES IN GRAPHS Nickolay Khadzhiivanov, Nedyalko Nenov Communicated by V. Drensky Abstract. Let Γ(M) where M V (G) be the set of all vertices

More information

On the independence number of graphs with maximum degree 3

On the independence number of graphs with maximum degree 3 On the independence number of graphs with maximum degree 3 Iyad A. Kanj Fenghui Zhang Abstract Let G be an undirected graph with maximum degree at most 3 such that G does not contain any of the three graphs

More information

The degree, size and chromatic index of a uniform hypergraph

The degree, size and chromatic index of a uniform hypergraph The degree, size and chromatic index of a uniform hypergraph Noga Alon Jeong Han Kim Abstract Let H be a k-uniform hypergraph in which no two edges share more than t common vertices, and let D denote the

More information

On-line Ramsey Theory for Bounded Degree Graphs

On-line Ramsey Theory for Bounded Degree Graphs On-line Ramsey Theory for Bounded Degree Graphs Jane Butterfield Tracy Grauman William B. Kinnersley Kevin G. Milans Christopher Stocker Douglas B. West University of Illinois Urbana IL, U.S.A. Submitted:

More information

Ramsey numbers for bipartite graphs with small bandwidth

Ramsey numbers for bipartite graphs with small bandwidth Ramsey numbers for bipartite graphs with small bandwidth Guilherme O. Mota 1,, Gábor N. Sárközy 2,, Mathias Schacht 3,, and Anusch Taraz 4, 1 Instituto de Matemática e Estatística, Universidade de São

More information

Generalized Induced Factor Problems

Generalized Induced Factor Problems Egerváry Research Group on Combinatorial Optimization Technical reports TR-2002-07. Published by the Egrerváry Research Group, Pázmány P. sétány 1/C, H 1117, Budapest, Hungary. Web site: www.cs.elte.hu/egres.

More information

On the k-path cover problem for cacti

On the k-path cover problem for cacti On the k-path cover problem for cacti Zemin Jin and Xueliang Li Center for Combinatorics and LPMC Nankai University Tianjin 300071, P.R. China zeminjin@eyou.com, x.li@eyou.com Abstract In this paper we

More information

Induced subgraphs of Ramsey graphs with many distinct degrees

Induced subgraphs of Ramsey graphs with many distinct degrees Journal of Combinatorial Theory, Series B 97 (2007) 62 69 www.elsevier.com/locate/jctb Induced subgraphs of Ramsey graphs with many distinct degrees Boris Bukh a, Benny Sudakov a,b a Department of Mathematics,

More information

SHARP BOUNDS FOR THE SUM OF THE SQUARES OF THE DEGREES OF A GRAPH

SHARP BOUNDS FOR THE SUM OF THE SQUARES OF THE DEGREES OF A GRAPH 31 Kragujevac J. Math. 25 (2003) 31 49. SHARP BOUNDS FOR THE SUM OF THE SQUARES OF THE DEGREES OF A GRAPH Kinkar Ch. Das Department of Mathematics, Indian Institute of Technology, Kharagpur 721302, W.B.,

More information

Graphical degree sequences and realizations

Graphical degree sequences and realizations swap Graphical and realizations Péter L. Erdös Alfréd Rényi Institute of Mathematics Hungarian Academy of Sciences MAPCON 12 MPIPKS - Dresden, May 15, 2012 swap Graphical and realizations Péter L. Erdös

More information

On the Relationship between Classes P and NP

On the Relationship between Classes P and NP Journal of Computer Science 8 (7): 1036-1040, 2012 ISSN 1549-3636 2012 Science Publications On the Relationship between Classes P and NP Anatoly D. Plotnikov Department of Computer Systems and Networks,

More information

arxiv:1112.0829v1 [math.pr] 5 Dec 2011

arxiv:1112.0829v1 [math.pr] 5 Dec 2011 How Not to Win a Million Dollars: A Counterexample to a Conjecture of L. Breiman Thomas P. Hayes arxiv:1112.0829v1 [math.pr] 5 Dec 2011 Abstract Consider a gambling game in which we are allowed to repeatedly

More information

Introduced by Stuart Kauffman (ca. 1986) as a tunable family of fitness landscapes.

Introduced by Stuart Kauffman (ca. 1986) as a tunable family of fitness landscapes. 68 Part II. Combinatorial Models can require a number of spin flips that is exponential in N (A. Haken et al. ca. 1989), and that one can in fact embed arbitrary computations in the dynamics (Orponen 1995).

More information

Ph.D. Thesis. Judit Nagy-György. Supervisor: Péter Hajnal Associate Professor

Ph.D. Thesis. Judit Nagy-György. Supervisor: Péter Hajnal Associate Professor Online algorithms for combinatorial problems Ph.D. Thesis by Judit Nagy-György Supervisor: Péter Hajnal Associate Professor Doctoral School in Mathematics and Computer Science University of Szeged Bolyai

More information

Cycle transversals in bounded degree graphs

Cycle transversals in bounded degree graphs Electronic Notes in Discrete Mathematics 35 (2009) 189 195 www.elsevier.com/locate/endm Cycle transversals in bounded degree graphs M. Groshaus a,2,3 P. Hell b,3 S. Klein c,1,3 L. T. Nogueira d,1,3 F.

More information

Cycles in a Graph Whose Lengths Differ by One or Two

Cycles in a Graph Whose Lengths Differ by One or Two Cycles in a Graph Whose Lengths Differ by One or Two J. A. Bondy 1 and A. Vince 2 1 LABORATOIRE DE MATHÉMATIQUES DISCRÉTES UNIVERSITÉ CLAUDE-BERNARD LYON 1 69622 VILLEURBANNE, FRANCE 2 DEPARTMENT OF MATHEMATICS

More information

High degree graphs contain large-star factors

High degree graphs contain large-star factors High degree graphs contain large-star factors Dedicated to László Lovász, for his 60th birthday Noga Alon Nicholas Wormald Abstract We show that any finite simple graph with minimum degree d contains a

More information

A 2-factor in which each cycle has long length in claw-free graphs

A 2-factor in which each cycle has long length in claw-free graphs A -factor in which each cycle has long length in claw-free graphs Roman Čada Shuya Chiba Kiyoshi Yoshimoto 3 Department of Mathematics University of West Bohemia and Institute of Theoretical Computer Science

More information

Labeling outerplanar graphs with maximum degree three

Labeling outerplanar graphs with maximum degree three Labeling outerplanar graphs with maximum degree three Xiangwen Li 1 and Sanming Zhou 2 1 Department of Mathematics Huazhong Normal University, Wuhan 430079, China 2 Department of Mathematics and Statistics

More information

SOLUTIONS TO ASSIGNMENT 1 MATH 576

SOLUTIONS TO ASSIGNMENT 1 MATH 576 SOLUTIONS TO ASSIGNMENT 1 MATH 576 SOLUTIONS BY OLIVIER MARTIN 13 #5. Let T be the topology generated by A on X. We want to show T = J B J where B is the set of all topologies J on X with A J. This amounts

More information

Finding and counting given length cycles

Finding and counting given length cycles Finding and counting given length cycles Noga Alon Raphael Yuster Uri Zwick Abstract We present an assortment of methods for finding and counting simple cycles of a given length in directed and undirected

More information

CS 598CSC: Combinatorial Optimization Lecture date: 2/4/2010

CS 598CSC: Combinatorial Optimization Lecture date: 2/4/2010 CS 598CSC: Combinatorial Optimization Lecture date: /4/010 Instructor: Chandra Chekuri Scribe: David Morrison Gomory-Hu Trees (The work in this section closely follows [3]) Let G = (V, E) be an undirected

More information

Lecture 4: BK inequality 27th August and 6th September, 2007

Lecture 4: BK inequality 27th August and 6th September, 2007 CSL866: Percolation and Random Graphs IIT Delhi Amitabha Bagchi Scribe: Arindam Pal Lecture 4: BK inequality 27th August and 6th September, 2007 4. Preliminaries The FKG inequality allows us to lower bound

More information

Strong Ramsey Games: Drawing on an infinite board

Strong Ramsey Games: Drawing on an infinite board Strong Ramsey Games: Drawing on an infinite board Dan Hefetz Christopher Kusch Lothar Narins Alexey Pokrovskiy Clément Requilé Amir Sarid arxiv:1605.05443v2 [math.co] 25 May 2016 May 26, 2016 Abstract

More information

1 if 1 x 0 1 if 0 x 1

1 if 1 x 0 1 if 0 x 1 Chapter 3 Continuity In this chapter we begin by defining the fundamental notion of continuity for real valued functions of a single real variable. When trying to decide whether a given function is or

More information

JUST-IN-TIME SCHEDULING WITH PERIODIC TIME SLOTS. Received December May 12, 2003; revised February 5, 2004

JUST-IN-TIME SCHEDULING WITH PERIODIC TIME SLOTS. Received December May 12, 2003; revised February 5, 2004 Scientiae Mathematicae Japonicae Online, Vol. 10, (2004), 431 437 431 JUST-IN-TIME SCHEDULING WITH PERIODIC TIME SLOTS Ondřej Čepeka and Shao Chin Sung b Received December May 12, 2003; revised February

More information

Every tree contains a large induced subgraph with all degrees odd

Every tree contains a large induced subgraph with all degrees odd Every tree contains a large induced subgraph with all degrees odd A.J. Radcliffe Carnegie Mellon University, Pittsburgh, PA A.D. Scott Department of Pure Mathematics and Mathematical Statistics University

More information

The positive minimum degree game on sparse graphs

The positive minimum degree game on sparse graphs The positive minimum degree game on sparse graphs József Balogh Department of Mathematical Sciences University of Illinois, USA jobal@math.uiuc.edu András Pluhár Department of Computer Science University

More information

A threshold for the Maker-Breaker clique game

A threshold for the Maker-Breaker clique game A threshold for the Maker-Breaker clique game Tobias Müller Miloš Stojaković October 7, 01 Abstract We study the Maker-Breaker k-clique game played on the edge set of the random graph G(n, p. In this game,

More information

Definition 11.1. Given a graph G on n vertices, we define the following quantities:

Definition 11.1. Given a graph G on n vertices, we define the following quantities: Lecture 11 The Lovász ϑ Function 11.1 Perfect graphs We begin with some background on perfect graphs. graphs. First, we define some quantities on Definition 11.1. Given a graph G on n vertices, we define

More information

Approximated Distributed Minimum Vertex Cover Algorithms for Bounded Degree Graphs

Approximated Distributed Minimum Vertex Cover Algorithms for Bounded Degree Graphs Approximated Distributed Minimum Vertex Cover Algorithms for Bounded Degree Graphs Yong Zhang 1.2, Francis Y.L. Chin 2, and Hing-Fung Ting 2 1 College of Mathematics and Computer Science, Hebei University,

More information

ON THE COMPLEXITY OF THE GAME OF SET. {kamalika,pbg,dratajcz,hoeteck}@cs.berkeley.edu

ON THE COMPLEXITY OF THE GAME OF SET. {kamalika,pbg,dratajcz,hoeteck}@cs.berkeley.edu ON THE COMPLEXITY OF THE GAME OF SET KAMALIKA CHAUDHURI, BRIGHTEN GODFREY, DAVID RATAJCZAK, AND HOETECK WEE {kamalika,pbg,dratajcz,hoeteck}@cs.berkeley.edu ABSTRACT. Set R is a card game played with a

More information

Zachary Monaco Georgia College Olympic Coloring: Go For The Gold

Zachary Monaco Georgia College Olympic Coloring: Go For The Gold Zachary Monaco Georgia College Olympic Coloring: Go For The Gold Coloring the vertices or edges of a graph leads to a variety of interesting applications in graph theory These applications include various

More information

Generating models of a matched formula with a polynomial delay

Generating models of a matched formula with a polynomial delay Generating models of a matched formula with a polynomial delay Petr Savicky Institute of Computer Science, Academy of Sciences of Czech Republic, Pod Vodárenskou Věží 2, 182 07 Praha 8, Czech Republic

More information

Degree Hypergroupoids Associated with Hypergraphs

Degree Hypergroupoids Associated with Hypergraphs Filomat 8:1 (014), 119 19 DOI 10.98/FIL1401119F Published by Faculty of Sciences and Mathematics, University of Niš, Serbia Available at: http://www.pmf.ni.ac.rs/filomat Degree Hypergroupoids Associated

More information

On-line choosability. Grzegorz Gutowski

On-line choosability. Grzegorz Gutowski Theoretical Computer Science Department Faculty of Mathematics and Computer Science Jagiellonian University On-line choosability Grzegorz Gutowski grzegorz.gutowski@tcs.uj.edu.pl Ph.D. Thesis Adviser:

More information

Minimum degree condition forcing complete graph immersion

Minimum degree condition forcing complete graph immersion Minimum degree condition forcing complete graph immersion Matt DeVos Department of Mathematics Simon Fraser University Burnaby, B.C. V5A 1S6 Jacob Fox Department of Mathematics MIT Cambridge, MA 02139

More information

An inequality for the group chromatic number of a graph

An inequality for the group chromatic number of a graph An inequality for the group chromatic number of a graph Hong-Jian Lai 1, Xiangwen Li 2 and Gexin Yu 3 1 Department of Mathematics, West Virginia University Morgantown, WV 26505 USA 2 Department of Mathematics

More information

Discrete Mathematics & Mathematical Reasoning Chapter 10: Graphs

Discrete Mathematics & Mathematical Reasoning Chapter 10: Graphs Discrete Mathematics & Mathematical Reasoning Chapter 10: Graphs Kousha Etessami U. of Edinburgh, UK Kousha Etessami (U. of Edinburgh, UK) Discrete Mathematics (Chapter 6) 1 / 13 Overview Graphs and Graph

More information

CSC2420 Fall 2012: Algorithm Design, Analysis and Theory

CSC2420 Fall 2012: Algorithm Design, Analysis and Theory CSC2420 Fall 2012: Algorithm Design, Analysis and Theory Allan Borodin November 15, 2012; Lecture 10 1 / 27 Randomized online bipartite matching and the adwords problem. We briefly return to online algorithms

More information

On Integer Additive Set-Indexers of Graphs

On Integer Additive Set-Indexers of Graphs On Integer Additive Set-Indexers of Graphs arxiv:1312.7672v4 [math.co] 2 Mar 2014 N K Sudev and K A Germina Abstract A set-indexer of a graph G is an injective set-valued function f : V (G) 2 X such that

More information

Stationary random graphs on Z with prescribed iid degrees and finite mean connections

Stationary random graphs on Z with prescribed iid degrees and finite mean connections Stationary random graphs on Z with prescribed iid degrees and finite mean connections Maria Deijfen Johan Jonasson February 2006 Abstract Let F be a probability distribution with support on the non-negative

More information

CHAPTER II THE LIMIT OF A SEQUENCE OF NUMBERS DEFINITION OF THE NUMBER e.

CHAPTER II THE LIMIT OF A SEQUENCE OF NUMBERS DEFINITION OF THE NUMBER e. CHAPTER II THE LIMIT OF A SEQUENCE OF NUMBERS DEFINITION OF THE NUMBER e. This chapter contains the beginnings of the most important, and probably the most subtle, notion in mathematical analysis, i.e.,

More information

MA651 Topology. Lecture 6. Separation Axioms.

MA651 Topology. Lecture 6. Separation Axioms. MA651 Topology. Lecture 6. Separation Axioms. This text is based on the following books: Fundamental concepts of topology by Peter O Neil Elements of Mathematics: General Topology by Nicolas Bourbaki Counterexamples

More information

Basic Concepts of Point Set Topology Notes for OU course Math 4853 Spring 2011

Basic Concepts of Point Set Topology Notes for OU course Math 4853 Spring 2011 Basic Concepts of Point Set Topology Notes for OU course Math 4853 Spring 2011 A. Miller 1. Introduction. The definitions of metric space and topological space were developed in the early 1900 s, largely

More information

Why? A central concept in Computer Science. Algorithms are ubiquitous.

Why? A central concept in Computer Science. Algorithms are ubiquitous. Analysis of Algorithms: A Brief Introduction Why? A central concept in Computer Science. Algorithms are ubiquitous. Using the Internet (sending email, transferring files, use of search engines, online

More information

All trees contain a large induced subgraph having all degrees 1 (mod k)

All trees contain a large induced subgraph having all degrees 1 (mod k) All trees contain a large induced subgraph having all degrees 1 (mod k) David M. Berman, A.J. Radcliffe, A.D. Scott, Hong Wang, and Larry Wargo *Department of Mathematics University of New Orleans New

More information

Small Maximal Independent Sets and Faster Exact Graph Coloring

Small Maximal Independent Sets and Faster Exact Graph Coloring Small Maximal Independent Sets and Faster Exact Graph Coloring David Eppstein Univ. of California, Irvine Dept. of Information and Computer Science The Exact Graph Coloring Problem: Given an undirected

More information

The chromatic spectrum of mixed hypergraphs

The chromatic spectrum of mixed hypergraphs The chromatic spectrum of mixed hypergraphs Tao Jiang, Dhruv Mubayi, Zsolt Tuza, Vitaly Voloshin, Douglas B. West March 30, 2003 Abstract A mixed hypergraph is a triple H = (X, C, D), where X is the vertex

More information

An example of a computable

An example of a computable An example of a computable absolutely normal number Verónica Becher Santiago Figueira Abstract The first example of an absolutely normal number was given by Sierpinski in 96, twenty years before the concept

More information

CMSC 858T: Randomized Algorithms Spring 2003 Handout 8: The Local Lemma

CMSC 858T: Randomized Algorithms Spring 2003 Handout 8: The Local Lemma CMSC 858T: Randomized Algorithms Spring 2003 Handout 8: The Local Lemma Please Note: The references at the end are given for extra reading if you are interested in exploring these ideas further. You are

More information

Short Cycles make W-hard problems hard: FPT algorithms for W-hard Problems in Graphs with no short Cycles

Short Cycles make W-hard problems hard: FPT algorithms for W-hard Problems in Graphs with no short Cycles Short Cycles make W-hard problems hard: FPT algorithms for W-hard Problems in Graphs with no short Cycles Venkatesh Raman and Saket Saurabh The Institute of Mathematical Sciences, Chennai 600 113. {vraman

More information

INDISTINGUISHABILITY OF ABSOLUTELY CONTINUOUS AND SINGULAR DISTRIBUTIONS

INDISTINGUISHABILITY OF ABSOLUTELY CONTINUOUS AND SINGULAR DISTRIBUTIONS INDISTINGUISHABILITY OF ABSOLUTELY CONTINUOUS AND SINGULAR DISTRIBUTIONS STEVEN P. LALLEY AND ANDREW NOBEL Abstract. It is shown that there are no consistent decision rules for the hypothesis testing problem

More information

A Sublinear Bipartiteness Tester for Bounded Degree Graphs

A Sublinear Bipartiteness Tester for Bounded Degree Graphs A Sublinear Bipartiteness Tester for Bounded Degree Graphs Oded Goldreich Dana Ron February 5, 1998 Abstract We present a sublinear-time algorithm for testing whether a bounded degree graph is bipartite

More information

Offline 1-Minesweeper is NP-complete

Offline 1-Minesweeper is NP-complete Offline 1-Minesweeper is NP-complete James D. Fix Brandon McPhail May 24 Abstract We use Minesweeper to illustrate NP-completeness proofs, arguments that establish the hardness of solving certain problems.

More information

A REMARK ON ALMOST MOORE DIGRAPHS OF DEGREE THREE. 1. Introduction and Preliminaries

A REMARK ON ALMOST MOORE DIGRAPHS OF DEGREE THREE. 1. Introduction and Preliminaries Acta Math. Univ. Comenianae Vol. LXVI, 2(1997), pp. 285 291 285 A REMARK ON ALMOST MOORE DIGRAPHS OF DEGREE THREE E. T. BASKORO, M. MILLER and J. ŠIRÁŇ Abstract. It is well known that Moore digraphs do

More information

Tree-representation of set families and applications to combinatorial decompositions

Tree-representation of set families and applications to combinatorial decompositions Tree-representation of set families and applications to combinatorial decompositions Binh-Minh Bui-Xuan a, Michel Habib b Michaël Rao c a Department of Informatics, University of Bergen, Norway. buixuan@ii.uib.no

More information

Combinatorial 5/6-approximation of Max Cut in graphs of maximum degree 3

Combinatorial 5/6-approximation of Max Cut in graphs of maximum degree 3 Combinatorial 5/6-approximation of Max Cut in graphs of maximum degree 3 Cristina Bazgan a and Zsolt Tuza b,c,d a LAMSADE, Université Paris-Dauphine, Place du Marechal de Lattre de Tassigny, F-75775 Paris

More information

Clique coloring B 1 -EPG graphs

Clique coloring B 1 -EPG graphs Clique coloring B 1 -EPG graphs Flavia Bonomo a,c, María Pía Mazzoleni b,c, and Maya Stein d a Departamento de Computación, FCEN-UBA, Buenos Aires, Argentina. b Departamento de Matemática, FCE-UNLP, La

More information

OHJ-2306 Introduction to Theoretical Computer Science, Fall 2012 8.11.2012

OHJ-2306 Introduction to Theoretical Computer Science, Fall 2012 8.11.2012 276 The P vs. NP problem is a major unsolved problem in computer science It is one of the seven Millennium Prize Problems selected by the Clay Mathematics Institute to carry a $ 1,000,000 prize for the

More information

Mathematics for Computer Science/Software Engineering. Notes for the course MSM1F3 Dr. R. A. Wilson

Mathematics for Computer Science/Software Engineering. Notes for the course MSM1F3 Dr. R. A. Wilson Mathematics for Computer Science/Software Engineering Notes for the course MSM1F3 Dr. R. A. Wilson October 1996 Chapter 1 Logic Lecture no. 1. We introduce the concept of a proposition, which is a statement

More information

Adaptive Online Gradient Descent

Adaptive Online Gradient Descent Adaptive Online Gradient Descent Peter L Bartlett Division of Computer Science Department of Statistics UC Berkeley Berkeley, CA 94709 bartlett@csberkeleyedu Elad Hazan IBM Almaden Research Center 650

More information

Odd induced subgraphs in graphs of maximum degree three

Odd induced subgraphs in graphs of maximum degree three Odd induced subgraphs in graphs of maximum degree three David M. Berman, Hong Wang, and Larry Wargo Department of Mathematics University of New Orleans New Orleans, Louisiana, USA 70148 Abstract A long-standing

More information

An Approximation Algorithm for Bounded Degree Deletion

An Approximation Algorithm for Bounded Degree Deletion An Approximation Algorithm for Bounded Degree Deletion Tomáš Ebenlendr Petr Kolman Jiří Sgall Abstract Bounded Degree Deletion is the following generalization of Vertex Cover. Given an undirected graph

More information

Game Chromatic Index of Graphs with Given Restrictions on Degrees

Game Chromatic Index of Graphs with Given Restrictions on Degrees Game Chromatic Index of Graphs with Given Restrictions on Degrees Andrew Beveridge Department of Mathematics and Computer Science Macalester College St. Paul, MN 55105 Tom Bohman, Alan Frieze, and Oleg

More information

Mathematics for Algorithm and System Analysis

Mathematics for Algorithm and System Analysis Mathematics for Algorithm and System Analysis for students of computer and computational science Edward A. Bender S. Gill Williamson c Edward A. Bender & S. Gill Williamson 2005. All rights reserved. Preface

More information

1 Definitions. Supplementary Material for: Digraphs. Concept graphs

1 Definitions. Supplementary Material for: Digraphs. Concept graphs Supplementary Material for: van Rooij, I., Evans, P., Müller, M., Gedge, J. & Wareham, T. (2008). Identifying Sources of Intractability in Cognitive Models: An Illustration using Analogical Structure Mapping.

More information

Embedding nearly-spanning bounded degree trees

Embedding nearly-spanning bounded degree trees Embedding nearly-spanning bounded degree trees Noga Alon Michael Krivelevich Benny Sudakov Abstract We derive a sufficient condition for a sparse graph G on n vertices to contain a copy of a tree T of

More information

Fairness in Routing and Load Balancing

Fairness in Routing and Load Balancing Fairness in Routing and Load Balancing Jon Kleinberg Yuval Rabani Éva Tardos Abstract We consider the issue of network routing subject to explicit fairness conditions. The optimization of fairness criteria

More information

Chapter 11. 11.1 Load Balancing. Approximation Algorithms. Load Balancing. Load Balancing on 2 Machines. Load Balancing: Greedy Scheduling

Chapter 11. 11.1 Load Balancing. Approximation Algorithms. Load Balancing. Load Balancing on 2 Machines. Load Balancing: Greedy Scheduling Approximation Algorithms Chapter Approximation Algorithms Q. Suppose I need to solve an NP-hard problem. What should I do? A. Theory says you're unlikely to find a poly-time algorithm. Must sacrifice one

More information

Applied Algorithm Design Lecture 5

Applied Algorithm Design Lecture 5 Applied Algorithm Design Lecture 5 Pietro Michiardi Eurecom Pietro Michiardi (Eurecom) Applied Algorithm Design Lecture 5 1 / 86 Approximation Algorithms Pietro Michiardi (Eurecom) Applied Algorithm Design

More information

Approximation Algorithms

Approximation Algorithms Approximation Algorithms or: How I Learned to Stop Worrying and Deal with NP-Completeness Ong Jit Sheng, Jonathan (A0073924B) March, 2012 Overview Key Results (I) General techniques: Greedy algorithms

More information

Minimum Bisection is NP-hard on Unit Disk Graphs

Minimum Bisection is NP-hard on Unit Disk Graphs Minimum Bisection is NP-hard on Unit Disk Graphs Josep Díaz 1 and George B. Mertzios 2 1 Departament de Llenguatges i Sistemes Informátics, Universitat Politécnica de Catalunya, Spain. 2 School of Engineering

More information

Non-Separable Detachments of Graphs

Non-Separable Detachments of Graphs Egerváry Research Group on Combinatorial Optimization Technical reports TR-2001-12. Published by the Egrerváry Research Group, Pázmány P. sétány 1/C, H 1117, Budapest, Hungary. Web site: www.cs.elte.hu/egres.

More information

Single machine parallel batch scheduling with unbounded capacity

Single machine parallel batch scheduling with unbounded capacity Workshop on Combinatorics and Graph Theory 21th, April, 2006 Nankai University Single machine parallel batch scheduling with unbounded capacity Yuan Jinjiang Department of mathematics, Zhengzhou University

More information

Local periods and binary partial words: An algorithm

Local periods and binary partial words: An algorithm Local periods and binary partial words: An algorithm F. Blanchet-Sadri and Ajay Chriscoe Department of Mathematical Sciences University of North Carolina P.O. Box 26170 Greensboro, NC 27402 6170, USA E-mail:

More information

ON DEGREES IN THE HASSE DIAGRAM OF THE STRONG BRUHAT ORDER

ON DEGREES IN THE HASSE DIAGRAM OF THE STRONG BRUHAT ORDER Séminaire Lotharingien de Combinatoire 53 (2006), Article B53g ON DEGREES IN THE HASSE DIAGRAM OF THE STRONG BRUHAT ORDER RON M. ADIN AND YUVAL ROICHMAN Abstract. For a permutation π in the symmetric group

More information

An inequality for the group chromatic number of a graph

An inequality for the group chromatic number of a graph Discrete Mathematics 307 (2007) 3076 3080 www.elsevier.com/locate/disc Note An inequality for the group chromatic number of a graph Hong-Jian Lai a, Xiangwen Li b,,1, Gexin Yu c a Department of Mathematics,

More information

The Union-Find Problem Kruskal s algorithm for finding an MST presented us with a problem in data-structure design. As we looked at each edge,

The Union-Find Problem Kruskal s algorithm for finding an MST presented us with a problem in data-structure design. As we looked at each edge, The Union-Find Problem Kruskal s algorithm for finding an MST presented us with a problem in data-structure design. As we looked at each edge, cheapest first, we had to determine whether its two endpoints

More information

CONTRIBUTIONS TO ZERO SUM PROBLEMS

CONTRIBUTIONS TO ZERO SUM PROBLEMS CONTRIBUTIONS TO ZERO SUM PROBLEMS S. D. ADHIKARI, Y. G. CHEN, J. B. FRIEDLANDER, S. V. KONYAGIN AND F. PAPPALARDI Abstract. A prototype of zero sum theorems, the well known theorem of Erdős, Ginzburg

More information

Total colorings of planar graphs with small maximum degree

Total colorings of planar graphs with small maximum degree Total colorings of planar graphs with small maximum degree Bing Wang 1,, Jian-Liang Wu, Si-Feng Tian 1 Department of Mathematics, Zaozhuang University, Shandong, 77160, China School of Mathematics, Shandong

More information

The Goldberg Rao Algorithm for the Maximum Flow Problem

The Goldberg Rao Algorithm for the Maximum Flow Problem The Goldberg Rao Algorithm for the Maximum Flow Problem COS 528 class notes October 18, 2006 Scribe: Dávid Papp Main idea: use of the blocking flow paradigm to achieve essentially O(min{m 2/3, n 1/2 }

More information

Continued Fractions and the Euclidean Algorithm

Continued Fractions and the Euclidean Algorithm Continued Fractions and the Euclidean Algorithm Lecture notes prepared for MATH 326, Spring 997 Department of Mathematics and Statistics University at Albany William F Hammond Table of Contents Introduction

More information

COMBINATORIAL PROPERTIES OF THE HIGMAN-SIMS GRAPH. 1. Introduction

COMBINATORIAL PROPERTIES OF THE HIGMAN-SIMS GRAPH. 1. Introduction COMBINATORIAL PROPERTIES OF THE HIGMAN-SIMS GRAPH ZACHARY ABEL 1. Introduction In this survey we discuss properties of the Higman-Sims graph, which has 100 vertices, 1100 edges, and is 22 regular. In fact

More information

Online and Offline Selling in Limit Order Markets

Online and Offline Selling in Limit Order Markets Online and Offline Selling in Limit Order Markets Kevin L. Chang 1 and Aaron Johnson 2 1 Yahoo Inc. klchang@yahoo-inc.com 2 Yale University ajohnson@cs.yale.edu Abstract. Completely automated electronic

More information

! Solve problem to optimality. ! Solve problem in poly-time. ! Solve arbitrary instances of the problem. !-approximation algorithm.

! Solve problem to optimality. ! Solve problem in poly-time. ! Solve arbitrary instances of the problem. !-approximation algorithm. Approximation Algorithms Chapter Approximation Algorithms Q Suppose I need to solve an NP-hard problem What should I do? A Theory says you're unlikely to find a poly-time algorithm Must sacrifice one of

More information

Bicolored Shortest Paths in Graphs with Applications to Network Overlay Design

Bicolored Shortest Paths in Graphs with Applications to Network Overlay Design Bicolored Shortest Paths in Graphs with Applications to Network Overlay Design Hongsik Choi and Hyeong-Ah Choi Department of Electrical Engineering and Computer Science George Washington University Washington,

More information

CSC 373: Algorithm Design and Analysis Lecture 16

CSC 373: Algorithm Design and Analysis Lecture 16 CSC 373: Algorithm Design and Analysis Lecture 16 Allan Borodin February 25, 2013 Some materials are from Stephen Cook s IIT talk and Keven Wayne s slides. 1 / 17 Announcements and Outline Announcements

More information

SHORT CYCLE COVERS OF GRAPHS WITH MINIMUM DEGREE THREE

SHORT CYCLE COVERS OF GRAPHS WITH MINIMUM DEGREE THREE SHOT YLE OVES OF PHS WITH MINIMUM DEEE THEE TOMÁŠ KISE, DNIEL KÁL, END LIDIKÝ, PVEL NEJEDLÝ OET ŠÁML, ND bstract. The Shortest ycle over onjecture of lon and Tarsi asserts that the edges of every bridgeless

More information

Analysis of Algorithms, I

Analysis of Algorithms, I Analysis of Algorithms, I CSOR W4231.002 Eleni Drinea Computer Science Department Columbia University Thursday, February 26, 2015 Outline 1 Recap 2 Representing graphs 3 Breadth-first search (BFS) 4 Applications

More information