hw5solutions

hw5solutions - c 1 In the above derivation A=I,b=u,D=our...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
ORF 307 Homework 5 Solutions Exercise 1. The adjacency matrix A (including the extra link) is: 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 1 - 1 0 1 - 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 - 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 - 1 0 0 - 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 1 0 0 - 1 0 - 1 0 1 0 - 1 0 - 1 0 0 0 0 0 0 0 0 - 1 1 0 0 0 0 0 0 0 - 1 1 0 0 0 0 0 0 0 0 0 0 0 - 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 - 1 - 1 - 1 0 0 0 1 0 0 0 0 0 0 0 0 - 1 0 0 0 0 0 1 1 0 0 0 The max flow problem is the same as we’ve done: min - t s.t. A T x - et = 0 0 x u Where [A -e] is the above adjacency matrix and u is a vector of capacities given in the problem. See the Matlab file for how to solve this problem. One solution to the problem is: x = (4 . 0 , 7 . 0 , 4 . 0 , 0 . 0 , 4 . 4 , 0 . 0 , 2 . 0 , 2 . 4 , 3 . 7 , 3 . 0 , 1 . 0 , 0 . 2 , 0 . 7 , 7 . 2 , 3 . 0 , 0 . 7 , 2 . 5 , 1 . 2 , 11 . 0) The dual variables associated with the flow constraints are: z = (0 , 0 , 1 , 1 , 0 , 1 , 1 , 1 , 1 , 1) The dual is derived as such: The Lagrangian of this problem is: L ( x, λ 1 , λ 2 , ν ) = c T x + λ T 1 ( Ax - b ) - λ T 2 x + ν T Dx with λ 1 , λ 2 0 The Lagrange dual function is: g ( λ 1 , λ 2 , ν ) = inf x±D L ( x, λ 1 , λ 2 , ν ) = inf x±D c T x + λ T 1 ( Ax - b ) - λ T 2 x + ν T Dx 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
= inf x±D ( c T + λ T 1 A - λ T 2 + ν T D ) x - λ T 1 b = - λ T 1 b if c + A T λ 1 - λ 2 + D T ν = 0, = -∞ otherwise Thus the dual problem is: max g ( λ 1 , λ 2 , ν ) s.t. λ 1 , λ 2 0 This is equivalent to: max - λ T 1 b s.t. c + A T λ 1 - λ 2 + D T ν = 0 λ 1 , λ 2 0 Note that λ 2 is like a surplus variable and we switch max to min so this problem is equivalent to: min λ T 1 b s.t. A T λ 1 + D T ν ≥ -
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: c 1 In the above derivation A=I,b=u,D=our adjacency matrix above, and c=[0,. ..,0,-1]. Not-ing the structure in D (every column has one 1 and one -1), we can write our problem as: min y T u s.t. y 9 , 1 + z 9-z 1 1 y i,j + z i-z j y Using the multipliers given above from Matlab, our minimum cut is U=1,2,5 and V=3,4,6,7,8,9. Notice that the cut could have gone through arc 2 or through both arcs 5 and 17. 2...
View Full Document

This note was uploaded on 04/30/2008 for the course ORF 307 taught by Professor Alexandrew.d'aspremont during the Spring '08 term at Princeton.

Page1 / 2

hw5solutions - c 1 In the above derivation A=I,b=u,D=our...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online