This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 6.896 Sublinear Time Algorithms September 23, 2004 Lecture 4 Lecturer: Ronitt Rubinfeld Scribe: Seth Gilbert Overview • Approximating the weight of the minimum spanning tree – Assume from last lecture : approx num cc( G, ) approximates the number of connected components in an n node graph, G , with degree ≤ d in time O ( d/ 4 ). – This time : Determine the (approximate) weight of a minimum spanning tree (MST) for an n node graph, G , with degree ≤ d . – Plan : Characterize minimum spanning tree in terms of number of connected components of certain subgraphs. Use approx num cc to estimate these values. • Testing Bipartiteness – Adjacency Matrix Model : The adjacency matrix model is defined. – Problem Statement : The problem of testing for bipartitness is presented. – Sketch : We sketch the algorithm to test for bipartitness and its proof. 1 Approximating the weight of the minimum spanning tree 1.1 Problem Statement Assume you are given: • a graph, G = ( V, E ), of n nodes and m edges with degree ≤ d , • G is connected, • each edge ( i, j ) ∈ E is assigned an integer weight w ij ∈ { 1 , . . . , w } , and • G is represented as an adjacency list augmented with weights for each edge. We define the minimum spanning tree (MST) weight as follows: MST ( G ) = min T spans G w ( T ) = X ( i,j ) ∈ T w ij Our goal is to output d MST ( G ) such that: (1 − ) MST ( G ) ≤ d MST ( G ) ≤ (1 + ) MST ( G ) Comments • Notice that the resulting algorithm provides no indication of how to actually construct the minimum spanning tree. • Notice that n − 1 ≤ MST ( G ), since each edge has weight at least 1, and there are n − 1 edges in a minimum spanning tree....
View
Full Document
 Spring '04
 RonittRubinfeld
 Algorithms, Graph Theory, edges, Bipartite graph

Click to edit the document details