Loading...
Please wait, while we are loading the content...
Similar Documents
Csc5160: combinatorial optimization and approximation algorithms topic: perfect matching polytope date: 22/02/2008.
| Content Provider | CiteSeerX |
|---|---|
| Author | Wu, Xiaobing Chan, Hei Ding, Ling Lau, Chi Lap, Lecturer Yuk, Scribe |
| Abstract | In this lecture, the focus is on general perfect matching problem where the goal is to prove that it can be solved in polynomial time by linear programming. Based on the LP formulation for bipartite matching studied in Lecture 10, we add some valid inequalities to establish a new formulation. Then we prove that for general perfect matching, all vertex solutions of the linear program are integral. This shows that the new formulation defines the matching polytope, and the general perfect matching problem can be solved by linear programming. Finally, we will show that this linear program can be solved in polynomial time, by providing a polynomial time separation oracle. 12.1 Formulation of General Perfect Matching Given a weighted general graph G = (V, E), for any edge e, we represents the weight of the edge and xe is the indicator variable of e. If an edge e is in the matching, we set xe = 1, otherwise xe = 0. Base on these definitions, the weighted perfect matching problem can be formulated as follows: max � e∈δ(v) e∈E |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | General Perfect Matching Given Combinatorial Optimization Linear Program Perfect Matching Polytope Date General Perfect Matching Problem Polynomial Time Linear Programming Polynomial Time Separation Oracle General Perfect Indicator Variable Valid Inequality Weighted General Graph Approximation Algorithm Topic Lp Formulation General Perfect Matching New Formulation Vertex Solution Weighted Perfect |
| Content Type | Text |