Searcharxiv⌕ Search

arXiv subjects

Dirk Oliver Theis

Publications and source records attributed to Dirk Oliver Theis.

42 records · Page 3Linked to original sources
arXiv2008

The Virtual Private Network Design Problem with Concave Costs (Oberwolfach abstract)

The symmetric Virtual Private Network Design (VPND) problem is concerned with buying capacity on links (edges) in a communication network such that certain traffic demands can be met. We investigate a natural generalization of VPND where the cost per unit of capacity may decrease if a larger amount of capacity is reserved (economies of scale principle). The growth of the cost of capacity is modelled by a non-decreasing concave function $f$. We call the problem the concave symmetric Virtual Private Network Design (cVPND) problem. After showing that a generalization of the so-called Pyramidal Routing problem and hence also the cVPND have the so-called tree routing property, we study approximation algorithms for cVPND. For general $f$, using known results on the so-called Single Source Buy at Bulk problem by Grandoni and Italiano, we give a randomized 24.92-approximation algorithm.

math.OC↗
arXiv2008

The VPN Tree Routing Conjecture for Outerplanar Networks

The VPN Tree Routing Conjecture is a conjecture about the Virtual Private Network Design problem. It states that the symmetric version of the problem always has an optimum solution which has a tree-like structure. In recent work, Hurkens, Keijsper and Stougie (Proc. IPCO XI, 2005; SIAM J. Discrete Math., 2007) have shown that the conjecture holds when the network is a ring. A shorter proof of the VPN Conjecture for rings was found a few months ago by Grandoni, Kaibel, Oriolo and Skutella (to appear in Oper. Res. Lett., 2008). In their paper, Grandoni et al. introduce another conjecture, called the Pyramidal Routing Conjecture (or simply PR Conjecture), which implies the VPN Conjecture. Here we consider a strengthened version of the PR Conjecture. First we establish several general tools which can be applied in arbitrary networks. Then we use them to prove that outerplanar networks satisfy the PR Conjecture.

math.OC↗
arXiv2008

The Cops and Robber game on graphs with forbidden (induced) subgraphs

The two-player, complete information game of Cops and Robber is played on undirected finite graphs. A number of cops and one robber are positioned on vertices and take turns in sliding along edges. The cops win if, after a move, a cop and the robber are on the same vertex. The minimum number of cops needed to catch the robber on a graph is called the cop number of that graph. In this paper, we study the cop number in the classes of graphs defined by forbidding one or more graphs as either subgraphs or induced subgraphs. In the case of a single forbidden graph we completely characterize (for both relations) the graphs which force bounded cop number. En passant, we bound the cop number in terms of tree-width.

math.CO↗
arXiv2008

The Cops & Robber game on series-parallel graphs

The Cops and Robber game is played on undirected finite graphs. $k$ cops and one robber are positioned on vertices and take turn in moving along edges. The cops win if, after a move, a cop and the robber are on the same vertex. A graph is called $k$-copwin, if the cops have a winning strategy. It is known that planar graphs are 3-copwin (Aigner & Fromme, 1984) and that outerplanar graphs are 2-copwin (Clarke, 2002). In this short note, we prove that series-parallel (i.e., graphs with no $K_4$ minor) graphs are 2-copwin. It is a well-known trick in the literature of cops & robber games to define variants of the game which impose restrictions on the possible strategies of the cops (see Clarke, 2002). For our proof, we define the ``cops & robber game with exits''. Our proof yields a winning strategy for the cops.

math.CO↗
arXiv2006

Odd minimum cut sets and b-matchings revisited

The famous Padberg-Rao separation algorithm for b-matching polyhedra can be implemented to run in O(n^2m log(n^2/m)) time in the uncapacitated case, and in O(nm^2 log(n^2/m)) time in the capacitated case (where n is the number of vertices and m is the number of edges of the underlying graph). We give a new and simple algorithm for the capacitated case which can be implemented to run in O(n^2m log(n^2/m)) time.

math.OC↗