Table of Contents
CSE 589 Part III
Dynamic Programming Summary
PPT Slide
PPT Slide
PPT Slide
Readings
Maximum Flow
Max-flow outline:
Properties of Flow: f(v,w) -- flow on edge (v,w)
An augmenting path with respect to a given flow f is a
Using an augmenting path to increase flow
Augmenting Path Theorem: A flow f is maximum iff it admits no augmenting path
=> Celebrated Max-flow Min-Cut Theorem
Residual Graph w.r.t. flow f
Ford-Fulkerson Method (G,s,t)
Edmonds-Karp
The shortest path distance from v to t in Rf is non-decreasing.
Shortest paths non-decreasing, cont.
Lemma: between any two consecutive saturations of (v,w), both d(v) and d(w) increase by at least 2.
=> Running time of Edmonds-Karp is O(m2n)
PPT Slide
PPT Slide
Fastest max-flow algorithms: preflow-push
Some applications of max-flow and max-flow min-cut theorem
Scheduling on uniform parallel machines
PPT Slide
PPT Slide
Bipartite Matching
Network Connectivity
Video on Demand
PPT Slide
Other network flow problems: 1. With lower bounds on flow.
Other network flow problems: 2. Minimum flow
Other network flow problems: 3. Min-cost max-flow
Classical application: Transportation Problem
Example: Fiat makes Uno and Ferrari’s.
Disk head scheduling
PPT Slide
|
Author: Anna Karlin
Email: karlin@cs.washington.edu
Home Page: http://www.cs.washington.edu/education/courses/589
Download presentation source
|