Locating Sources to Meet Flow Demands in Undirected Networks 1
|
|
- Leo Cannon
- 5 years ago
- Views:
Transcription
1 Journal of Algorithms 42, (2002) doi: /jagm ,available online at on Locating Sources to Meet Flow Demands in Undirected Networks 1 Kouji Arata Matsushita Communication Industrial Co., Ltd., Yokohama , Japan Satoru Iwata Department ofmathematical Informatics, Graduate School ofinformation Science and Technology, University oftokyo, Tokyo , Japan iwata@sr3.t.u-tokyo.ac.jp Kazuhisa Makino Division ofsystems Science, Graduate School ofengineering Science, Osaka University, Toyonaka, Osaka , Japan makino@sys.es.osaka-u.ac.jp and Satoru Fujishige Division ofsystems Science, Graduate School ofengineering Science, Osaka University, Toyonaka, Osaka , Japan fujishig@sys.es.osaka-u.ac.jp Received March 8,2000 This paper deals with the problem of finding a minimum-cost vertex subset S in an undirected network such that for each vertex v we can send d v units of flow from S to v. Although this problem is NP-hard in general,h. Tamura,H. Sugawara, M. Sengoku,and S. Shinoda (IEICE Trans. Fund. J81-A (1998), ) have 1 A preliminary version of this paper appears in Proceedings of 7th Scandinavian Workshop on Algorithm Theory (SWAT2000) [2]. The authors gratefully acknowledge the partial support of a Scientific Grant in Aid from the Ministry of Education,Science,Sports and Culture of Japan. This work was done while K. Arata and S. Iwata were with the Division of Systems Science,Graduate School of Engineering Science,Osaka University /02 $ Elsevier Science All rights reserved. 54
2 source location in networks 55 presented a greedy algorithm for solving the special case with uniform costs on the vertices. We give a simpler proof on the validity of the greedy algorithm using linear programming duality and improve the running time bound from O n 2 M n m to O nm n m,where n is the number of vertices in the network and M n m denotes the time for max-flow computation in a network with n vertices and m edges. We also present an O n m + n log n time algorithm for the special case with uniform demands and arbitrary costs Elsevier Science Key Words: location problem; edge-connectivity; max-flow; greedy algorithm; maximum adjacency ordering. 1. INTRODUCTION Let = G u d c be an undirected network on the underlying graph G = V E with a vertex set V of cardinality n and an edge set E of cardinality m. It has a capacity function u E R +,a demand function d V R +,and a cost function c V R +,where R + denotes the set of nonnegative reals. This paper addresses the problem of finding a minimumcost vertex subset S V such that the maximum flow value between S and v is at least d v for each v V. For a pair of disjoint subsets X Y V,we denote by λ X Y the maximum flow value between X and Y in. For any v V we write the singleton set v as v. For convenience,we assign λ X Y =+ if X Y. Note that λ is independent of the demand function d. Now,the problem is formulated as follows. Minimize c v v S subject to S V (1) λ S v d v v V We call this problem the Source Location problem. We say that a vertex set S V covers a vertex v if λ S v d v. Namely, Source Location asks for a minimum-cost subset S V that covers all the vertices in V. A motivation of this problem is given as follows. Suppose we are asked to locate multiple servers S which can provide a certain service in the multimedia network. A user at a vertex v can receive the service by connecting to a server in S through a path in. To ensure the quality of the service to v even if d v links become out of order,we should select S so that the edge-connectivity between S and v is at least d v. Thus the problem of finding a minimum-cost fault-tolerant setting of servers is an instance of Source Location with unit edge capacities. A special case of this problem with a uniform cost function was introduced by Tamura et al. [12,13]. They called it the plural cover problem. They
3 56 arata et al. first considered the case in which both d and c are uniform and described an algorithm that runs in O n 2 M n m time [12],where M n m denotes the time complexity of computing an s-t maximum flow in an undirected network with n vertices and m edges (see Goldberg and Rao [5,6] for recent development after the book of Ahuja et al. [1] was published). Later Tamura et al. [13] showed that a simple greedy algorithm solves problem Source Location in O n 2 M n m time even if the demand function d is arbitrary while the cost function c is still uniform. Ito et al. [7] described another algorithm to improve the time complexity to O npm n m,where p is the number of distinct values of d v v V ; i.e., p = d v v V. In this paper,we analyze the greedy algorithm of Tamura et al. [13] to give a simpler proof based on the linear programming duality. We then improve the greedy algorithm to run in O nm n m time. In the case of uniform demand and arbitrary cost,we give an O n m + n log n time algorithm. The algorithm makes use of maximum adjacency (MA) ordering,which was used by Nagamochi and Ibaraki for the minimum cut problem [8] and the edge-connectivity augmentation problem [9] for undirected networks. Finally,we show that Source Location is in general NP-hard,by reducing the knapsack problem to Source Location. Thus,it remains open to prove the NP-hardness in the strong sense or to devise a pseudo-polynomial time algorithm. We summarize the time complexity of Source Location in Table I, where bold letters indicate the complexity results obtained in this paper. The rest of the paper is organized as follows. Section 2 formulates Source Location as an integer programming problem,section 3 considers Source Location when c is uniform but d is arbitrary,and Section 4 discusses the case in which d is uniform but c is arbitrary. In Section 5,we show that Source Location is in general NP-hard. TABLE I Summary of the Results on Source Location c: uniform c: arbitrary d: uniform O n 2 M n m * Tamura et al. [12] O nm n m Ito et al. [7] O n m + n log n O n m + n log n d: arbitrary O n 2 M n m Tamura et al. [13] O npm n m ** Ito et al. [7] NP-hard O nm n m *M n m : the time complexity for computing a maximum s-t flow. **p : the number of distinct values of d v v V.
4 source location in networks INTEGER PROGRAMMING FORMULATION In this section,we formulate Source Location as an integer programming problem with an exponential number of constraints. A cut is a proper nonempty subset of V.ForacutX,let X denote the set of edges that cross X; i.e., X = e e = v w E v X w V X,and let κ X denote its capacity; i.e., κ X = u e e X If X = v,we write κ v instead of κ v. For a pair of disjoint vertex subsets X Y V,we denote κ X Y = e X Y u e. Forv V,we simply write κ X v instead of κ X v. We also denote by d W the maximum demand in W ; i.e., d W =max d v v W We say that a vertex v W attains the maximum demand in W if d v =d W. A cut W is called deficient if κ W <d W. IfacutW is deficient but any other nonempty subset X W is not, W is called a minimal deficient set. It is easy to see that the max-flow min-cut theorem implies the following lemma. Lemma 2.1 [13]. Let = G = V E u d c be an undirected network. Then S V covers all vertices in V if and only if S W holds for every minimal deficient set W. Let = W 1 W 2 W l be the family of all minimal deficient sets and let V = v 1 v 2 v n. Define an l n matrix A = A ij by A ij = 1if v j W i and A ij = 0 otherwise. From Lemma 2.1, Source Location can be written as the 0 1 integer programming problem, Minimize subject to n c j x j j=1 n A ij x j 1 i = 1 2 l (2) j=1 x j 0 1 j = 1 2 n where c j = c v j j = 1 2 n,and the 0 1 vector x = x 1 x 2 x n is the characteristic vector of a subset of V.
5 58 arata et al. Remark 2.1. Note that l = might be exponential in n and m. For example,let us consider a network = G = V E u d c,where V = v 1 v 2 v n, E = v 1 v i i = 2 3 n, u e =1 e E, { n if i = 1 d v i = 2 0 otherwise, and c is an arbitrary cost function. Then we can see that { } n = W W = + 1 W v 2 1 and hence = ( ) n 1 n 2 3. THE UNIFORM COST CASE 3.1. A Greedy Algorithm In this section,we consider Source Location with a uniform cost function. Specifically we assume that c v =1 v V. Tamura et al. [13] proposed the following greedy algorithm to solve Source Location. Algorithm Greedy Step 0. Order the vertices v 1 v 2 v n in V such that d v 1 d v 2 d v n. Step 1. Initialize j = 1 and S = V. Step 2. If S v j covers all vertices in V,then S = S v j. Step 3. If j = n then output S and halt. Otherwise, j = j + 1 and go to Step 2. Example 3.1. Let us apply Greedy to the network = G = V E u d c given in Fig. 1,where the values of u e (e E) and d v (v V ) are,respectively,attached to edges and vertices in Fig. 1 and c v =1 for all v V. The results are illustrated in Fig. 2. We initially include all vertices in S and check whether the set S v 1 covers all vertices or not. Since it covers v 1 (and hence all vertices in V ),we update S = S v 1 (see (i) in Fig. 2). We next check whether S v 2 covers all vertices or not. Since it covers both v 1 and v 2,we update S = S v 2 (see (ii)). By repeating this argument to the current S = v 3 v 4 v 7 (see (iii) (vii)),we finally obtain S = v 4 v 5 v 7 shown in (viii).
6 source location in networks 59 FIG. 1. A network = G = V E u d c,where c v =1 v V. FIG. 2. Algorithm Greedy applied to the network in Fig. 1.
7 60 arata et al. In order to show the correctness of algorithm Greedy,we consider the linear programming relaxation of (2), n Minimize c j x j and its dual, subject to Maximize subject to j=1 n A ij x j 1 i = 1 2 l (3) j=1 x j 0 l y i i=1 j = 1 2 n l A ij y i c j j = 1 2 n (4) i=1 y i 0 i = 1 2 l Recall that c j = 1 j = 1 2 n is assumed in this section. We also replace Steps 1 and 2 in algorithm Greedy as follows. Step 1. Initialize j = 1, S = V,and y i = 0 for i = 1 2 l. Step 2. (2-1) If S v j covers all vertices in V,then S = S v j. (2-2) Otherwise,choose a W i with W i S = v j,and set y i = 1. Obviously,the revised algorithm always keeps a feasible solution S (i.e., S covers all vertices in V ). Assume that Step 2 finds a vertex w that is not covered with S v j in some execution of Step 2 and Step 3. Then by the max-flow min-cut theorem,there exists a W i such that W i contains w and W i S v j =. Since S covers w,we have W i S = v j. Therefore,Step 2 always finds a W i with W i S = v j if S v j does not cover a vertex in V. Moreover,since S contains v l, l = j j + 1 n, j is the maximum number k with v k W i ; i.e., d v j =d W i. Let us also note that we need not actually perform Step 1 (initialization of y) in the revised version. It suffices to keep only non-zero y i s in Step 2. Let x and y be the primal and dual variables obtained at the end of the revised greedy algorithm. Note that x is the characteristic vector of the output S of the algorithm. The algorithm does not delete v j from S if and only if it updates y i as y i = 1 for some i with W i S = v j. Hence,at the termination,we have n l x j = (5) j=1 yi i=1
8 source location in networks 61 Therefore, x is a 0 1 solution satisfying (3) and (5). By the weak duality of linear programming problems (3) and (4),we only need to prove the feasibility of y in (4) to show the correctness of the algorithm Greedy. The feasibility of y will be proved by Lemmas 3.1 and 3.3 given below. Recall that the cut capacity function κ satisfies κ X +κ Y κ X Y +κ Y X X Y V (6) We call a set function satisfying (6) posi-modular [10]. The following lemma was originally proved by Tamura et al. [13]. Here we give a simple proof of it. Lemma 3.1 [13]. Let = G = V E u d c be an undirected flow network. Let W 1 and W 2 be minimal deficient sets in, and for each i = 1 2, let v i W i be a vertex that attains the maximum demand in W i.ifw 1 W 2, then we have v 1 W 1 W 2 or v 2 W 1 W 2. Proof. Suppose,to the contrary,that both v 1 W 1 W 2 and v 2 W 2 W 1 hold. Since W 1 and W 2 are deficient sets, d v 1 >κ W 1 and d v 2 > κ W 2 hold. It follows from (6) that d v 1 +d v 2 >κ W 1 +κ W 2 κ W 1 W 2 +κ W 2 W 1 This means that d v 1 >κ W 1 W 2 or d v 2 >κ W 2 W 1 holds. Since we have v 1 W 1 W 2 and v 2 W 2 W 1 by the assumption,it follows that W 1 W 2 or W 2 W 1 is deficient,which contradicts the minimality of W 1 or W 2. Arrange the columns of A in such a way that d v 1 d v 2 d v n. For each index i with 1 i l,let k i denote the maximum number k with v k W i. Then Lemma 3.1 implies the following. Lemma 3.2. as a submatrix. The matrix A does not contain ( j k i 1 k i 2 ) i i (7) Proof. If A contains such a submatrix,we have W i1 W i2, k i 1 W i1 W i2,and k i 2 W i2 W i1,which contradicts Lemma 3.1. Lemma 3.3. The dual variables y obtained by the revised greedy algorithm give a feasible solution to (4).
9 62 arata et al. Proof. Let S be the solution obtained by the revised algorithm. Suppose that for a pair of distinct rows i 1 i 2 we have yi 1 = yi 2 = 1. It follows from the discussion after Steps 1 and 2 that W il S = k i l l = 1 2 (8) Note that k i 1 k i 2 holds,since otherwise yi 1 and yi 2 must be updated in the same iteration in Step 2,a contradiction. Hence,from (8) and Lemma 3.2,we have W i1 W i2 =. We have thus shown the following. Theorem 3.1. If the cost function c is uniform, then algorithm Greedy produces an optimal solution of Source Location An Efficient Implementation We now analyze the time complexity of algorithm Greedy. Steps 0,1, and 3 are clearly executed in O n log n, O 1,and O n time,respectively. As for Step 2,Tamura et al. [13] checked if S v j covers all vertices in V by computing λ S v j v i (i.e.,a max flow from S v j to v i ) for all v i. Clearly,this requires O nm n m time. Since Step 2 is iterated n times,the required time is O n 2 M n m in total [13]. However,the following lemma implies that Step 2 can be replaced by Step 2. If S v j covers v j,then S = S v j. Lemma 3.4. If S v j covers v j in Step 2 of algorithm Greedy, then S v j covers v i for all i j. Proof. Assume that some v i with i<jis not covered with S v j. Then there exists a cut X with v i X, V X S v j,and κ X < d v i. Then, S X v j clearly holds. Moreover,we have S X = v j since otherwise S does not cover v i,which contradicts the property that Greedy always keeps a feasible set S. Hence, X separates v j and S v j. Since κ X <d v i d v j,it follows that S v j does not cover v j,a contradiction. Thus we have improved the time complexity by use of Step 2 instead of Step 2. Theorem 3.2. If the cost function c is uniform, then problem Source Location can be solved in O nm n m time.
10 source location in networks THE UNIFORM DEMAND CASE In this section,we consider Source Location with a uniform demand function d. We assume that d v =g (a fixed positive real) holds for all v V. We show that,in this setting,source Location can be solved in O n m + n log n time. Our algorithm uses the maximum adjacency ordering and does not use any maximum flow computations. An ordering v 1 v 2 v n of all vertices in V is called a maximum adjacency (MA) ordering if it satisfies κ v 1 v 2 v i v i+1 κ v 1 v 2 v i v j for 1 i<j n The MA ordering plays a crucial role in this section through the following lemma. Lemma 4.1 [8, 11]. Let G = V E be an undirected graph with a nonnegative capacity function u. Then, the following statements hold. (i) An MA ordering v 1 v 2 v n can be computed in O m + n log n time. (ii) The last two vertices v n 1 and v n for every MA ordering in G satisfy λ v n 1 v n =κ v n (9) We can choose the first vertex v 1 arbitrarily. If the demand function d is uniform,minimal deficient sets are pairwise disjoint by the posi-modularity 6 of κ; i.e.,for every pair of W 1 and W 2 in, W 1 W 2 = holds. Therefore,in order to solve Source Location,we try to find all minimal deficient sets W and construct a minimum-cost source set S V by choosing from each W a vertex v W with the minimum cost c v among W. Since any source set S must contain all v V such that κ v <g,we initialize S as S = v V κ v <g. If S n 1,then we output S and halt. Otherwise,if S,we construct a network from by merging all vertices in S into a single vertex s,and if S =,we construct a network from by adding a new isolated vertex s. Then we compute an MA ordering v 1 = s v 2 v h 1 v h starting from s in. Denote by λ and κ,respectively,the maximum flow value and the cut capacity in. Since all vertices v except for s in satisfy κ v g,lemma 4.1 implies λ v h 1 v h =κ v h g
11 64 arata et al. Namely,every cut X in that separates v h 1 and v h satisfies κ X g. We claim that no minimal deficient set W in separates v h 1 and v h. Assume to the contrary that there exists such a minimal deficient set W. Then W contains a vertex w S,since otherwise κ W =κ W g a contradiction. Then it follows from w and the minimality of W that W is given as W = w. However,this does not separate v h 1 and v h, which contradicts the assumption. From our claim,we can update by merging the vertices v h 1 and v h into a single vertex ˆv. We then check if ˆv satisfies κ ˆv g. Since κ ˆv <gimplies that W = v h 1 v h is a minimal deficient set in the original network,if κ ˆv <g,we again update the network by merging s and ˆv into s and update S by adding v h 1 if c v h 1 <c v h ; otherwise,we add v h to S. Now we have κ ˆv g for all vertices except for s in. By repeating the above argument for (i.e.,we compute an MA ordering v 1 = s v 2 v h 1 v h in,merge the last two vertices v h 1 and v h,and so on),we can compute a minimum-cost source set S. Formally it can be written as follows. Algorithm Contract Input. A network = G = V E u d c,where d v = g for all v. Output. A minimum-cost vertex set S V which covers all vertices in V. Step 0. Initialize S = v V κ v <g and α v =v for all v V. Step 1. If S V 1,then output S and halt. Otherwise,construct a network from by merging all vertices in S into a single vertex s. Step 2. (2-I) Compute an MA ordering v 1 = s v 2 v h 1 v h starting from s in. (2-II) Merge the last two vertices v h 1 and v h into a single vertex ˆv. Denote the resulting network by again. (2-III) If c α v h 1 <c α v h,then α ˆv =α v h 1 ; otherwise, α ˆv =α v h. (2-IV) If κ ˆv <g,then update by merging s and ˆv into s, and S = S α ˆv. Step 3. If V 2,where V denotes the vertex set of,then output S and halt. Otherwise go to Step2.
12 source location in networks 65 FIG. 3. A network = G = V E u d c,where d v =8 v V. Note that the algorithm uses α for computing,from each W,a vertex v W with the minimum cost c v among W. Formally, α v with v s stores the vertex v in the original network having the minimum cost c v among P v,where P v is the set of all vertices v in V which are merged to v. Example 4.1. Let us apply algorithm Contract to the network = G = V E u d c given in Fig. 3,where u e (e E) and c v (v V ) are,respectively,attached to edges and vertices in Fig. 3 and d v =8 for all v. The results are illustrated in Fig. 4. In Step 0,the algorithm initializes S = v b and α (see (i) in Fig. 4),and since S < V 1,Step 1 updates the network (i) to (ii). Step 2 then computes an MA ordering v 1 = s v 2 = v e v 3 = v d v 4 = v a v 5 = v c (see (iii-1)) and merges v 4 = v a and v 5 = v c into v ac (see (iii-2)). Since c α v a <c α v c, Step 2 puts α v ac =α v a. Moreover,by κ v ac < 8,Step 2 updates by merging s and v ac into s,and S = S α v ac = v a (see (iii-3)). Since V =3,the algorithm returns to Step 2. Step 2 again computes an MA ordering v 1 = s v 2 = v d v 3 = v e (see (iv-1)) and merges v 2 = v d and v 3 = v e into v de (see (iv-2)). Since c α v d <c α v e,step 2 puts α v de =α v d. Moreover,by κ v de < 8,Step 2 updates by merging s and v de into s,and S = S α v de = v d (see (iv-3)). Now,since V =1,Step 3 outputs S = v a v b v d whose cost is 6. Theorem 4.1. Problem Source Location can be solved in O n m + n log n time if the demand function d is uniform. Proof. Since the above discussion shows the correctness of algorithm Contract,we only consider its time complexity. Clearly Steps 0,1,and 3 take O n + m time. Step 2 can be executed in O n m + n log n time since it has n 1 iterations and each iteration takes O m + n log n time by Lemma 4.1. Therefore,in total,it requires O n m + n log n time. In the uniform demand case,minimal deficient sets are so-called extreme sets explored by Watanabe and Nakamura [14]. An extreme set is a vertex subset X such that κ Y >κ X for any proper nonempty subset Y X. Any efficient algorithm for finding all the extreme sets can be used to solve our problem. In particular,a recent randomized algorithm of Benczúr and
13 66 arata et al. FIG. 4. Algorithm Contract applied to the network in Fig. 3. Karger [3] leads to an Õ n2 algorithm that solves our problem with high probability. 5. NP-HARDNESS OF GENERAL CASE In this section,we show the NP-hardness of Source Location with nonuniform cost and demand functions. Theorem 5.1. Problem Source Location is NP-hard, even if the undirected graph G = V E is a star; i.e., E = v w w V \ v for some v V. Proof. We transform the NP-hard problem Knapsack to this problem [4]. Problem. Knapsack.
14 source location in networks 67 Input. A finite set Z = z 1 z 2 z n associated with a size function σ Z Z + and a value function τ Z Z + and a positive integer β z i Z σ z i,where Z + denotes the set of all nonnegative integers. Output. A subset X Z that is an optimal solution of Maximize τ z subject to z X σ z β (10) z X X Z It is easy to see that Knapsack is polynomially equivalent to the problem of computing a subset Y Z that solves Minimize τ z z Y subject to σ z σ z β z Y z Z (11) Y Z by identifying Y with Z X. Therefore,in the following we consider (11) instead of (10). For this problem instance,we consider an undirected network = G = V E u d c with V = Z z 0, E = z 0 z i z i Z, u z 0 z i = σ z i for i = 1 2 n,and σ z i β if i = 0 d z i = z i Z 0 otherwise, τ z i +1 if i = 0 c z i = z i Z τ z i otherwise. Note that d z i =0 for all z i Z. Therefore S V covers all vertices in V if and only if it covers z 0 ; i.e., λ S z 0 d z 0 = σ z i β (12) Moreover,since z 0 and Z covers z 0 and since c z 0 > z i Z c z i,an optimal S is contained in Z. This implies that λ S z 0 = u z 0 z i = σ z i (13) z i S z i S and hence (12) is equivalent to the constraint in (11). Since c z i =τ z i for all z i Z, S Z is an optimal solution for the instance of problem (11) if and only if it is optimal for the corresponding instance for Source Location. z i Z
15 68 arata et al. 6. CONCLUSION In this paper,we have analyzed the greedy algorithm of Tamura et al. [13] for Source Location with a uniform cost function and given a simpler proof based on the linear programming duality. We have also improved the Greedy algorithm to run in O nm n m time. Moreover,we have given an O n m + n log n time algorithm for Source Location with a uniform demand function. Finally,we have shown that Source Location is in general NP-hard by reducing Knapsack to Source Location. Thus, it remains open to prove the NP-hardness in the strong sense or to devise a pseudo-polynomial time algorithm. ACKNOWLEDGMENTS The authors thank the anonymous referees for their helpful and constructive comments, which improved the presentation of this paper. In particular,one of the reviewers suggested us a simpler version of algorithm Contract. REFERENCES 1. R. K. Ahuja,T. L. Magnanti,and J. B. Orlin, Network Flows, Prentice Hall,Englewood Cliffs,NJ., K. Arata,S. Iwata,K. Makino,and S. Fujishige,Locating sources to meet flow demands in undirected networks, in SWAT2000 (M. M. Halldorsson,Ed.),Springer Lecture Notes in Computer Science,Vol. 1851,pp ,Springer-Verlag,Berlin/New York, A. A. Benczúr and D. R. Karger,Augmenting undirected edge connectivity in Õ n2 time, J. Algorithms, 37 (2000), M. R. Garey and D. S. Johnson, Computers and Intractability, Freeman,New York, A. V. Goldberg and S. Rao,Beyond the flow decomposition barrier,j. ACM 45 (1998), A. V. Goldberg and S. Rao,Flows in undirected unit capacity networks,siam J. Discrete Math. 12 (1999), H. Ito,H. Uehara,and M. Yokoyama,A faster and flexible algorithm for a location problem on undirected flow networks, IEICE Trans. Fund. E83-A (2000), H. Nagamochi and T. Ibaraki,Computing edge-connectivity of multigraphs and capacitated graphs, SIAM J. Discrete Math. 5 (1992), H. Nagamochi and T. Ibaraki,Deterministic Õ nm time edge-splitting in undirected graphs, J. Combin. Optim. 1 (1997), H. Nagamochi and T. Ibaraki,A note on minimizing submodular functions,inform. Process. Lett. 67 (1998), M. Stoer and F. Wagner,A simple min cut algorithm,j. ACM 44 (1997) H. Tamura,M. Sengoku,S. Shinoda,and T. Abe,Some covering problems in location theory on flow networks, IEICE Trans. Fund. E75-A (1992), H. Tamura,H. Sugawara,M. Sengoku,and S. Shinoda,Plural cover problem on undirected flow networks (in Japanese), IEICE Trans. Fund. J81-A (1998), T. Watanabe and A. Nakamura,Edge connectivity augmentation problems,j. Comput. System Sci. 53 (1987),
The Multi-Commodity Source Location Problems and the Price of Greed
Journal of Graph Algorithms and Applications http://jgaa.info/ vol. 3, no., pp. 55 73 (29) The Multi-Commodity Source Location Problems and the Price of Greed Hiro Ito Mike Paterson 2 Kenya Sugihara Graduate
More informationRealization of set functions as cut functions of graphs and hypergraphs
Discrete Mathematics 226 (2001) 199 210 www.elsevier.com/locate/disc Realization of set functions as cut functions of graphs and hypergraphs Satoru Fujishige a;, Sachin B. Patkar b a Division of Systems
More informationThe Maximum Flow Problem with Disjunctive Constraints
The Maximum Flow Problem with Disjunctive Constraints Ulrich Pferschy Joachim Schauer Abstract We study the maximum flow problem subject to binary disjunctive constraints in a directed graph: A negative
More informationDual Consistent Systems of Linear Inequalities and Cardinality Constrained Polytopes. Satoru FUJISHIGE and Jens MASSBERG.
RIMS-1734 Dual Consistent Systems of Linear Inequalities and Cardinality Constrained Polytopes By Satoru FUJISHIGE and Jens MASSBERG December 2011 RESEARCH INSTITUTE FOR MATHEMATICAL SCIENCES KYOTO UNIVERSITY,
More informationA necessary and sufficient condition for the existence of a spanning tree with specified vertices having large degrees
A necessary and sufficient condition for the existence of a spanning tree with specified vertices having large degrees Yoshimi Egawa Department of Mathematical Information Science, Tokyo University of
More informationMAKING BIPARTITE GRAPHS DM-IRREDUCIBLE
SIAM J. DISCRETE MATH. Vol. 32, No. 1, pp. 560 590 c 2018 Society for Industrial and Applied Mathematics MAKING BIPARTITE GRAPHS DM-IRREDUCIBLE KRISTÓF BÉRCZI, SATORU IWATA, JUN KATO, AND YUTARO YAMAGUCHI
More informationAn algorithm to increase the node-connectivity of a digraph by one
Discrete Optimization 5 (2008) 677 684 Contents lists available at ScienceDirect Discrete Optimization journal homepage: www.elsevier.com/locate/disopt An algorithm to increase the node-connectivity of
More informationMinimum Cost Source Location Problems with Flow Requirements
Algorithmica (2008) 50: 555 583 DOI 10.1007/s00453-007-9012-y Minimum Cost Source Location Problems with Flow Requirements Mariko Sakashita Kazuhisa Makino Satoru Fujishige Received: 1 May 2006 / Accepted:
More informationLecture #21. c T x Ax b. maximize subject to
COMPSCI 330: Design and Analysis of Algorithms 11/11/2014 Lecture #21 Lecturer: Debmalya Panigrahi Scribe: Samuel Haney 1 Overview In this lecture, we discuss linear programming. We first show that the
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS. Dijkstra s Algorithm and L-concave Function Maximization
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Dijkstra s Algorithm and L-concave Function Maximization Kazuo MUROTA and Akiyoshi SHIOURA METR 2012 05 March 2012; revised May 2012 DEPARTMENT OF MATHEMATICAL
More informationA Min-Max Theorem for k-submodular Functions and Extreme Points of the Associated Polyhedra. Satoru FUJISHIGE and Shin-ichi TANIGAWA.
RIMS-1787 A Min-Max Theorem for k-submodular Functions and Extreme Points of the Associated Polyhedra By Satoru FUJISHIGE and Shin-ichi TANIGAWA September 2013 RESEARCH INSTITUTE FOR MATHEMATICAL SCIENCES
More informationThe minimum G c cut problem
The minimum G c cut problem Abstract In this paper we define and study the G c -cut problem. Given a complete undirected graph G = (V ; E) with V = n, edge weighted by w(v i, v j ) 0 and an undirected
More informationLecture 15 (Oct 6): LP Duality
CMPUT 675: Approximation Algorithms Fall 2014 Lecturer: Zachary Friggstad Lecture 15 (Oct 6): LP Duality Scribe: Zachary Friggstad 15.1 Introduction by Example Given a linear program and a feasible solution
More information1 Some loose ends from last time
Cornell University, Fall 2010 CS 6820: Algorithms Lecture notes: Kruskal s and Borůvka s MST algorithms September 20, 2010 1 Some loose ends from last time 1.1 A lemma concerning greedy algorithms and
More informationdirected weighted graphs as flow networks the Ford-Fulkerson algorithm termination and running time
Network Flow 1 The Maximum-Flow Problem directed weighted graphs as flow networks the Ford-Fulkerson algorithm termination and running time 2 Maximum Flows and Minimum Cuts flows and cuts max flow equals
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Combinatorial Relaxation Algorithm for the Entire Sequence of the Maximum Degree of Minors in Mixed Polynomial Matrices Shun SATO (Communicated by Taayasu MATSUO)
More informationFRACTIONAL PACKING OF T-JOINS. 1. Introduction
FRACTIONAL PACKING OF T-JOINS FRANCISCO BARAHONA Abstract Given a graph with nonnegative capacities on its edges, it is well known that the capacity of a minimum T -cut is equal to the value of a maximum
More informationTight Approximation Ratio of a General Greedy Splitting Algorithm for the Minimum k-way Cut Problem
Algorithmica (2011 59: 510 520 DOI 10.1007/s00453-009-9316-1 Tight Approximation Ratio of a General Greedy Splitting Algorithm for the Minimum k-way Cut Problem Mingyu Xiao Leizhen Cai Andrew Chi-Chih
More informationCapacitated Network Design on Undirected Graphs
Capacitated Network Design on Undirected Graphs Deeparnab Chakrabarty, Ravishankar Krishnaswamy, Shi Li, and Srivatsan Narayanan Abstract. In this paper, we study the approximability of the capacitated
More information1 Column Generation and the Cutting Stock Problem
1 Column Generation and the Cutting Stock Problem In the linear programming approach to the traveling salesman problem we used the cutting plane approach. The cutting plane approach is appropriate when
More informationAn Improved Approximation Algorithm for Maximum Edge 2-Coloring in Simple Graphs
An Improved Approximation Algorithm for Maximum Edge 2-Coloring in Simple Graphs Zhi-Zhong Chen Ruka Tanahashi Lusheng Wang Abstract We present a polynomial-time approximation algorithm for legally coloring
More informationPacking and Covering Dense Graphs
Packing and Covering Dense Graphs Noga Alon Yair Caro Raphael Yuster Abstract Let d be a positive integer. A graph G is called d-divisible if d divides the degree of each vertex of G. G is called nowhere
More informationFINAL EXAM PRACTICE PROBLEMS CMSC 451 (Spring 2016)
FINAL EXAM PRACTICE PROBLEMS CMSC 451 (Spring 2016) The final exam will be on Thursday, May 12, from 8:00 10:00 am, at our regular class location (CSI 2117). It will be closed-book and closed-notes, except
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS. Deductive Inference for the Interiors and Exteriors of Horn Theories
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Deductive Inference for the Interiors and Exteriors of Horn Theories Kazuhisa MAKINO and Hirotaka ONO METR 2007 06 February 2007 DEPARTMENT OF MATHEMATICAL INFORMATICS
More informationAlgorithms and Theory of Computation. Lecture 11: Network Flow
Algorithms and Theory of Computation Lecture 11: Network Flow Xiaohui Bei MAS 714 September 18, 2018 Nanyang Technological University MAS 714 September 18, 2018 1 / 26 Flow Network A flow network is a
More informationAugmenting Outerplanar Graphs to Meet Diameter Requirements
Proceedings of the Eighteenth Computing: The Australasian Theory Symposium (CATS 2012), Melbourne, Australia Augmenting Outerplanar Graphs to Meet Diameter Requirements Toshimasa Ishii Department of Information
More informationThe Steiner k-cut Problem
The Steiner k-cut Problem Chandra Chekuri Sudipto Guha Joseph (Seffi) Naor September 23, 2005 Abstract We consider the Steiner k-cut problem which generalizes both the k-cut problem and the multiway cut
More informationMaximum flow problem
Maximum flow problem 7000 Network flows Network Directed graph G = (V, E) Source node s V, sink node t V Edge capacities: cap : E R 0 Flow: f : E R 0 satisfying 1. Flow conservation constraints e:target(e)=v
More informationThe partial inverse minimum cut problem with L 1 -norm is strongly NP-hard. Elisabeth Gassner
FoSP Algorithmen & mathematische Modellierung FoSP Forschungsschwerpunkt Algorithmen und mathematische Modellierung The partial inverse minimum cut problem with L 1 -norm is strongly NP-hard Elisabeth
More informationExact Algorithms for Dominating Induced Matching Based on Graph Partition
Exact Algorithms for Dominating Induced Matching Based on Graph Partition Mingyu Xiao School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu 611731,
More informationAn Improved Algorithm for Parameterized Edge Dominating Set Problem
An Improved Algorithm for Parameterized Edge Dominating Set Problem Ken Iwaide and Hiroshi Nagamochi Department of Applied Mathematics and Physics, Graduate School of Informatics, Kyoto University, Japan,
More informationMulticommodity Flows and Column Generation
Lecture Notes Multicommodity Flows and Column Generation Marc Pfetsch Zuse Institute Berlin pfetsch@zib.de last change: 2/8/2006 Technische Universität Berlin Fakultät II, Institut für Mathematik WS 2006/07
More informationLecture 11 October 7, 2013
CS 4: Advanced Algorithms Fall 03 Prof. Jelani Nelson Lecture October 7, 03 Scribe: David Ding Overview In the last lecture we talked about set cover: Sets S,..., S m {,..., n}. S has cost c S. Goal: Cover
More informationDiscrete Applied Mathematics
Discrete Applied Mathematics 159 (2011) 1345 1351 Contents lists available at ScienceDirect Discrete Applied Mathematics journal homepage: www.elsevier.com/locate/dam On disconnected cuts and separators
More informationA Faster Strongly Polynomial Time Algorithm for Submodular Function Minimization
A Faster Strongly Polynomial Time Algorithm for Submodular Function Minimization James B. Orlin Sloan School of Management, MIT Cambridge, MA 02139 jorlin@mit.edu Abstract. We consider the problem of minimizing
More informationLecture 10 February 4, 2013
UBC CPSC 536N: Sparse Approximations Winter 2013 Prof Nick Harvey Lecture 10 February 4, 2013 Scribe: Alexandre Fréchette This lecture is about spanning trees and their polyhedral representation Throughout
More informationReduction of Ultrametric Minimum Cost Spanning Tree Games to Cost Allocation Games on Rooted Trees. Kazutoshi ANDO and Shinji KATO.
RIMS-1674 Reduction of Ultrametric Minimum Cost Spanning Tree Games to Cost Allocation Games on Rooted Trees By Kazutoshi ANDO and Shinji KATO June 2009 RESEARCH INSTITUTE FOR MATHEMATICAL SCIENCES KYOTO
More informationMinmax Tree Cover in the Euclidean Space
Journal of Graph Algorithms and Applications http://jgaa.info/ vol. 15, no. 3, pp. 345 371 (2011) Minmax Tree Cover in the Euclidean Space Seigo Karakawa 1 Ehab Morsy 1 Hiroshi Nagamochi 1 1 Department
More informationOn improving matchings in trees, via bounded-length augmentations 1
On improving matchings in trees, via bounded-length augmentations 1 Julien Bensmail a, Valentin Garnero a, Nicolas Nisse a a Université Côte d Azur, CNRS, Inria, I3S, France Abstract Due to a classical
More informationMatroid Representation of Clique Complexes
Matroid Representation of Clique Complexes Kenji Kashiwabara 1, Yoshio Okamoto 2, and Takeaki Uno 3 1 Department of Systems Science, Graduate School of Arts and Sciences, The University of Tokyo, 3 8 1,
More informationCS 6820 Fall 2014 Lectures, October 3-20, 2014
Analysis of Algorithms Linear Programming Notes CS 6820 Fall 2014 Lectures, October 3-20, 2014 1 Linear programming The linear programming (LP) problem is the following optimization problem. We are given
More informationDominating Set Counting in Graph Classes
Dominating Set Counting in Graph Classes Shuji Kijima 1, Yoshio Okamoto 2, and Takeaki Uno 3 1 Graduate School of Information Science and Electrical Engineering, Kyushu University, Japan kijima@inf.kyushu-u.ac.jp
More informationNATIONAL UNIVERSITY OF SINGAPORE CS3230 DESIGN AND ANALYSIS OF ALGORITHMS SEMESTER II: Time Allowed 2 Hours
NATIONAL UNIVERSITY OF SINGAPORE CS3230 DESIGN AND ANALYSIS OF ALGORITHMS SEMESTER II: 2017 2018 Time Allowed 2 Hours INSTRUCTIONS TO STUDENTS 1. This assessment consists of Eight (8) questions and comprises
More informationDownloaded 03/01/17 to Redistribution subject to SIAM license or copyright; see
SIAM J. DISCRETE MATH. Vol. 31, No. 1, pp. 335 382 c 2017 Society for Industrial and Applied Mathematics PARTITION CONSTRAINED COVERING OF A SYMMETRIC CROSSING SUPERMODULAR FUNCTION BY A GRAPH ATTILA BERNÁTH,
More informationMATHEMATICAL ENGINEERING TECHNICAL REPORTS. Induction of M-convex Functions by Linking Systems
MATHEMATICAL ENGINEERING TECHNICAL REPORTS Induction of M-convex Functions by Linking Systems Yusuke KOBAYASHI and Kazuo MUROTA METR 2006 43 July 2006 DEPARTMENT OF MATHEMATICAL INFORMATICS GRADUATE SCHOOL
More informationMinimization of Symmetric Submodular Functions under Hereditary Constraints
Minimization of Symmetric Submodular Functions under Hereditary Constraints J.A. Soto (joint work with M. Goemans) DIM, Univ. de Chile April 4th, 2012 1 of 21 Outline Background Minimal Minimizers and
More informationOn the adjacency matrix of a block graph
On the adjacency matrix of a block graph R. B. Bapat Stat-Math Unit Indian Statistical Institute, Delhi 7-SJSS Marg, New Delhi 110 016, India. email: rbb@isid.ac.in Souvik Roy Economics and Planning Unit
More informationCS264: Beyond Worst-Case Analysis Lecture #15: Smoothed Complexity and Pseudopolynomial-Time Algorithms
CS264: Beyond Worst-Case Analysis Lecture #15: Smoothed Complexity and Pseudopolynomial-Time Algorithms Tim Roughgarden November 5, 2014 1 Preamble Previous lectures on smoothed analysis sought a better
More informationOn Two Class-Constrained Versions of the Multiple Knapsack Problem
On Two Class-Constrained Versions of the Multiple Knapsack Problem Hadas Shachnai Tami Tamir Department of Computer Science The Technion, Haifa 32000, Israel Abstract We study two variants of the classic
More informationMA015: Graph Algorithms
MA015 3. Minimum cuts 1 MA015: Graph Algorithms 3. Minimum cuts (in undirected graphs) Jan Obdržálek obdrzalek@fi.muni.cz Faculty of Informatics, Masaryk University, Brno All pairs flows/cuts MA015 3.
More information1 T 1 = where 1 is the all-ones vector. For the upper bound, let v 1 be the eigenvector corresponding. u:(u,v) E v 1(u)
CME 305: Discrete Mathematics and Algorithms Instructor: Reza Zadeh (rezab@stanford.edu) Final Review Session 03/20/17 1. Let G = (V, E) be an unweighted, undirected graph. Let λ 1 be the maximum eigenvalue
More informationApproximating connectivity augmentation problems
Approximating connectivity augmentation problems Zeev Nutov The Open University of Israel Let G = (V, E) be an undirected graph and let S V. The S-connectivity λ S G (u, v) of a node pair (u, v) in G is
More informationReachability-based matroid-restricted packing of arborescences
Egerváry Research Group on Combinatorial Optimization Technical reports TR-2016-19. Published by the Egerváry Research Group, Pázmány P. sétány 1/C, H 1117, Budapest, Hungary. Web site: www.cs.elte.hu/egres.
More information6.854J / J Advanced Algorithms Fall 2008
MIT OpenCourseWare http://ocw.mit.edu 6.85J / 8.5J Advanced Algorithms Fall 008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. 8.5/6.85 Advanced Algorithms
More informationA Characterization of Graphs with Fractional Total Chromatic Number Equal to + 2
A Characterization of Graphs with Fractional Total Chromatic Number Equal to + Takehiro Ito a, William. S. Kennedy b, Bruce A. Reed c a Graduate School of Information Sciences, Tohoku University, Aoba-yama
More informationThe Restricted Edge-Connectivity of Kautz Undirected Graphs
The Restricted Edge-Connectivity of Kautz Undirected Graphs Ying-Mei Fan College of Mathematics and Information Science Guangxi University, Nanning, Guangxi, 530004, China Jun-Ming Xu Min Lü Department
More informationTopic: Primal-Dual Algorithms Date: We finished our discussion of randomized rounding and began talking about LP Duality.
CS787: Advanced Algorithms Scribe: Amanda Burton, Leah Kluegel Lecturer: Shuchi Chawla Topic: Primal-Dual Algorithms Date: 10-17-07 14.1 Last Time We finished our discussion of randomized rounding and
More informationOn Steepest Descent Algorithms for Discrete Convex Functions
On Steepest Descent Algorithms for Discrete Convex Functions Kazuo Murota Graduate School of Information Science and Technology, University of Tokyo, and PRESTO, JST E-mail: murota@mist.i.u-tokyo.ac.jp
More informationCS264: Beyond Worst-Case Analysis Lecture #18: Smoothed Complexity and Pseudopolynomial-Time Algorithms
CS264: Beyond Worst-Case Analysis Lecture #18: Smoothed Complexity and Pseudopolynomial-Time Algorithms Tim Roughgarden March 9, 2017 1 Preamble Our first lecture on smoothed analysis sought a better theoretical
More informationApproximating maximum satisfiable subsystems of linear equations of bounded width
Approximating maximum satisfiable subsystems of linear equations of bounded width Zeev Nutov The Open University of Israel Daniel Reichman The Open University of Israel Abstract We consider the problem
More informationA Polynomial-Time Algorithm for Pliable Index Coding
1 A Polynomial-Time Algorithm for Pliable Index Coding Linqi Song and Christina Fragouli arxiv:1610.06845v [cs.it] 9 Aug 017 Abstract In pliable index coding, we consider a server with m messages and n
More informationIntroduction to Bin Packing Problems
Introduction to Bin Packing Problems Fabio Furini March 13, 2015 Outline Origins and applications Applications: Definition: Bin Packing Problem (BPP) Solution techniques for the BPP Heuristic Algorithms
More informationMaking Bipartite Graphs DM-irreducible
Making Bipartite Graphs DM-irreducible Kristóf Bérczi 1, Satoru Iwata 2, Jun Kato 3, Yutaro Yamaguchi 4 1. Eötvös Lorand University, Hungary. 2. University of Tokyo, Japan. 3. TOYOTA Motor Corporation,
More informationApproximation algorithms for cycle packing problems
Approximation algorithms for cycle packing problems Michael Krivelevich Zeev Nutov Raphael Yuster Abstract The cycle packing number ν c (G) of a graph G is the maximum number of pairwise edgedisjoint cycles
More informationAn approximation algorithm for the partial covering 0 1 integer program
An approximation algorithm for the partial covering 0 1 integer program Yotaro Takazawa Shinji Mizuno Tomonari Kitahara submitted January, 2016 revised July, 2017 Abstract The partial covering 0 1 integer
More informationGeneralized Pigeonhole Properties of Graphs and Oriented Graphs
Europ. J. Combinatorics (2002) 23, 257 274 doi:10.1006/eujc.2002.0574 Available online at http://www.idealibrary.com on Generalized Pigeonhole Properties of Graphs and Oriented Graphs ANTHONY BONATO, PETER
More informationCS675: Convex and Combinatorial Optimization Fall 2014 Combinatorial Problems as Linear Programs. Instructor: Shaddin Dughmi
CS675: Convex and Combinatorial Optimization Fall 2014 Combinatorial Problems as Linear Programs Instructor: Shaddin Dughmi Outline 1 Introduction 2 Shortest Path 3 Algorithms for Single-Source Shortest
More informationarxiv: v1 [cs.ds] 26 Feb 2016
On the computational complexity of minimum-concave-cost flow in a two-dimensional grid Shabbir Ahmed, Qie He, Shi Li, George L. Nemhauser arxiv:1602.08515v1 [cs.ds] 26 Feb 2016 Abstract We study the minimum-concave-cost
More informationTwo Applications of Maximum Flow
Two Applications of Maximum Flow The Bipartite Matching Problem a bipartite graph as a flow network maximum flow and maximum matching alternating paths perfect matchings 2 Circulation with Demands flows
More informationLearning symmetric non-monotone submodular functions
Learning symmetric non-monotone submodular functions Maria-Florina Balcan Georgia Institute of Technology ninamf@cc.gatech.edu Nicholas J. A. Harvey University of British Columbia nickhar@cs.ubc.ca Satoru
More informationFall 2017 Qualifier Exam: OPTIMIZATION. September 18, 2017
Fall 2017 Qualifier Exam: OPTIMIZATION September 18, 2017 GENERAL INSTRUCTIONS: 1 Answer each question in a separate book 2 Indicate on the cover of each book the area of the exam, your code number, and
More informationLecture 21 November 11, 2014
CS 224: Advanced Algorithms Fall 2-14 Prof. Jelani Nelson Lecture 21 November 11, 2014 Scribe: Nithin Tumma 1 Overview In the previous lecture we finished covering the multiplicative weights method and
More informationNP-hardness of the stable matrix in unit interval family problem in discrete time
Systems & Control Letters 42 21 261 265 www.elsevier.com/locate/sysconle NP-hardness of the stable matrix in unit interval family problem in discrete time Alejandra Mercado, K.J. Ray Liu Electrical and
More informationMIT Algebraic techniques and semidefinite optimization February 14, Lecture 3
MI 6.97 Algebraic techniques and semidefinite optimization February 4, 6 Lecture 3 Lecturer: Pablo A. Parrilo Scribe: Pablo A. Parrilo In this lecture, we will discuss one of the most important applications
More informationDiscrete (and Continuous) Optimization WI4 131
Discrete (and Continuous) Optimization WI4 131 Kees Roos Technische Universiteit Delft Faculteit Electrotechniek, Wiskunde en Informatica Afdeling Informatie, Systemen en Algoritmiek e-mail: C.Roos@ewi.tudelft.nl
More informationOn a Cardinality-Constrained Transportation Problem With Market Choice
On a Cardinality-Constrained Transportation Problem With Market Choice Pelin Damcı-Kurt Santanu S. Dey Simge Küçükyavuz Abstract It is well-known that the intersection of the matching polytope with a cardinality
More informationFractional coloring and the odd Hadwiger s conjecture
European Journal of Combinatorics 29 (2008) 411 417 www.elsevier.com/locate/ejc Fractional coloring and the odd Hadwiger s conjecture Ken-ichi Kawarabayashi a, Bruce Reed b a National Institute of Informatics,
More informationAdvanced Combinatorial Optimization September 22, Lecture 4
8.48 Advanced Combinatorial Optimization September 22, 2009 Lecturer: Michel X. Goemans Lecture 4 Scribe: Yufei Zhao In this lecture, we discuss some results on edge coloring and also introduce the notion
More informationTORIC WEAK FANO VARIETIES ASSOCIATED TO BUILDING SETS
TORIC WEAK FANO VARIETIES ASSOCIATED TO BUILDING SETS YUSUKE SUYAMA Abstract. We give a necessary and sufficient condition for the nonsingular projective toric variety associated to a building set to be
More information1 Perfect Matching and Matching Polytopes
CS 598CSC: Combinatorial Optimization Lecture date: /16/009 Instructor: Chandra Chekuri Scribe: Vivek Srikumar 1 Perfect Matching and Matching Polytopes Let G = (V, E be a graph. For a set E E, let χ E
More informationOn shredders and vertex connectivity augmentation
On shredders and vertex connectivity augmentation Gilad Liberman The Open University of Israel giladliberman@gmail.com Zeev Nutov The Open University of Israel nutov@openu.ac.il Abstract We consider the
More informationRepresentations of All Solutions of Boolean Programming Problems
Representations of All Solutions of Boolean Programming Problems Utz-Uwe Haus and Carla Michini Institute for Operations Research Department of Mathematics ETH Zurich Rämistr. 101, 8092 Zürich, Switzerland
More informationEvery line graph of a 4-edge-connected graph is Z 3 -connected
European Journal of Combinatorics 0 (2009) 595 601 Contents lists available at ScienceDirect European Journal of Combinatorics journal homepage: www.elsevier.com/locate/ejc Every line graph of a 4-edge-connected
More informationAcyclic and Oriented Chromatic Numbers of Graphs
Acyclic and Oriented Chromatic Numbers of Graphs A. V. Kostochka Novosibirsk State University 630090, Novosibirsk, Russia X. Zhu Dept. of Applied Mathematics National Sun Yat-Sen University Kaohsiung,
More informationCS675: Convex and Combinatorial Optimization Fall 2016 Combinatorial Problems as Linear and Convex Programs. Instructor: Shaddin Dughmi
CS675: Convex and Combinatorial Optimization Fall 2016 Combinatorial Problems as Linear and Convex Programs Instructor: Shaddin Dughmi Outline 1 Introduction 2 Shortest Path 3 Algorithms for Single-Source
More informationEXERCISES SHORTEST PATHS: APPLICATIONS, OPTIMIZATION, VARIATIONS, AND SOLVING THE CONSTRAINED SHORTEST PATH PROBLEM. 1 Applications and Modelling
SHORTEST PATHS: APPLICATIONS, OPTIMIZATION, VARIATIONS, AND SOLVING THE CONSTRAINED SHORTEST PATH PROBLEM EXERCISES Prepared by Natashia Boland 1 and Irina Dumitrescu 2 1 Applications and Modelling 1.1
More informationComplexity Results for Equistable Graphs and Related Classes
Complexity Results for Equistable Graphs and Related Classes Martin Milanič James Orlin Gábor Rudolf January 20, 2009 Abstract We describe a simple pseudo-polynomial-time dynamic programming algorithm
More informationThe maximum edge-disjoint paths problem in complete graphs
Theoretical Computer Science 399 (2008) 128 140 www.elsevier.com/locate/tcs The maximum edge-disjoint paths problem in complete graphs Adrian Kosowski Department of Algorithms and System Modeling, Gdańsk
More informationK-center Hardness and Max-Coverage (Greedy)
IOE 691: Approximation Algorithms Date: 01/11/2017 Lecture Notes: -center Hardness and Max-Coverage (Greedy) Instructor: Viswanath Nagarajan Scribe: Sentao Miao 1 Overview In this lecture, we will talk
More informationFrom Primal-Dual to Cost Shares and Back: A Stronger LP Relaxation for the Steiner Forest Problem
From Primal-Dual to Cost Shares and Back: A Stronger LP Relaxation for the Steiner Forest Problem Jochen Könemann 1, Stefano Leonardi 2, Guido Schäfer 2, and Stefan van Zwam 3 1 Department of Combinatorics
More informationLecture 4: An FPTAS for Knapsack, and K-Center
Comp 260: Advanced Algorithms Tufts University, Spring 2016 Prof. Lenore Cowen Scribe: Eric Bailey Lecture 4: An FPTAS for Knapsack, and K-Center 1 Introduction Definition 1.0.1. The Knapsack problem (restated)
More informationApproximation Algorithms for Maximum. Coverage and Max Cut with Given Sizes of. Parts? A. A. Ageev and M. I. Sviridenko
Approximation Algorithms for Maximum Coverage and Max Cut with Given Sizes of Parts? A. A. Ageev and M. I. Sviridenko Sobolev Institute of Mathematics pr. Koptyuga 4, 630090, Novosibirsk, Russia fageev,svirg@math.nsc.ru
More informationLecture 4: NP and computational intractability
Chapter 4 Lecture 4: NP and computational intractability Listen to: Find the longest path, Daniel Barret What do we do today: polynomial time reduction NP, co-np and NP complete problems some examples
More informationA Characterization of (3+1)-Free Posets
Journal of Combinatorial Theory, Series A 93, 231241 (2001) doi:10.1006jcta.2000.3075, available online at http:www.idealibrary.com on A Characterization of (3+1)-Free Posets Mark Skandera Department of
More informationElisabeth Gassner Introduction THE PARTIAL INVERSE MINIMUM CUT PROBLEM WITH L 1 -NORM IS STRONGLY NP-HARD
RAIRO-Oper. Res. 44 (010) 41 49 DOI: 10.1051/ro/010017 RAIRO Operations Research www.rairo-ro.org THE PARTIAL INVERSE MINIMUM CUT PROBLEM WITH L 1 -NORM IS STRONGLY NP-HARD Elisabeth Gassner 1 Abstract.
More informationA Note on Submodular Function Minimization by Chubanov s LP Algorithm
A Note on Submodular Function Minimization by Chubanov s LP Algorithm Satoru FUJISHIGE September 25, 2017 Abstract Recently Dadush, Végh, and Zambelli (2017) has devised a polynomial submodular function
More informationThe Generalized Regenerator Location Problem
The Generalized Regenerator Location Problem Si Chen Ivana Ljubić S. Raghavan College of Business and Public Affairs, Murray State University Murray, KY 42071, USA Faculty of Business, Economics, and Statistics,
More informationAn Algorithmic Framework for Wireless Information Flow
An Algorithmic Framework for Wireless Information Flow The MIT Faculty has made this article openly available. Please share how this access benefits you. Your story matters. Citation As Published Publisher
More informationThe Reduction of Graph Families Closed under Contraction
The Reduction of Graph Families Closed under Contraction Paul A. Catlin, Department of Mathematics Wayne State University, Detroit MI 48202 November 24, 2004 Abstract Let S be a family of graphs. Suppose
More informationCS599: Convex and Combinatorial Optimization Fall 2013 Lecture 17: Combinatorial Problems as Linear Programs III. Instructor: Shaddin Dughmi
CS599: Convex and Combinatorial Optimization Fall 2013 Lecture 17: Combinatorial Problems as Linear Programs III Instructor: Shaddin Dughmi Announcements Today: Spanning Trees and Flows Flexibility awarded
More information