lecture20 - Duality Karush-Kuhn-Tucker and Lagrangian...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
April 25, 2006 IE447 – 1 Duality, Karush-Kuhn-Tucker and Lagrangian methods Aur´elie Thiele Lehigh University April 25, 2006 IE 417 Nonlinear Programming
Background image of page 1

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

View Full DocumentRight Arrow Icon
The Dual Problem Review on Duality Dual Problem Pros and Cons Favorable Structures April 25, 2006 IE447 – 2 s Consider the primal problem: min f ( x ) s.t. x X, g j ( x ) 0 , j = 1 , . . . , r, assuming it has a finite optimum.
Background image of page 2
The Dual Problem Review on Duality Dual Problem Pros and Cons Favorable Structures April 25, 2006 IE447 – 2 s Consider the primal problem: min f ( x ) s.t. x X, g j ( x ) 0 , j = 1 , . . . , r, assuming it has a finite optimum. s The dual problem is defined as: max q ( μ ) s.t. μ 0 , with dual function: q ( μ ) = inf x X { f ( x ) + μ g ( x ) b ² L Lagrangian }
Background image of page 3

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

View Full DocumentRight Arrow Icon
Pros and Cons for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 3
Background image of page 4
Pros for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 4 1. The dual is concave. By contrast, the primal may not be convex.
Background image of page 5

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

View Full DocumentRight Arrow Icon
Pros for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 4 1. The dual is concave. By contrast, the primal may not be convex. 2. The dual may have smaller dimension, and has simpler constraints.
Background image of page 6
Pros for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 4 1. The dual is concave. By contrast, the primal may not be convex. 2. The dual may have smaller dimension, and has simpler constraints. 3. If there is no duality gap and the dual is solved exactly for a Lagrange multiplier μ , all optimal primal solutions can be obtained by minimizing the Lagrangian L ( x, μ ) over x X .
Background image of page 7

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

View Full DocumentRight Arrow Icon
Pros for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 4 1. The dual is concave. By contrast, the primal may not be convex. 2. The dual may have smaller dimension, and has simpler constraints. 3. If there is no duality gap and the dual is solved exactly for a Lagrange multiplier μ , all optimal primal solutions can be obtained by minimizing the Lagrangian L ( x, μ ) over x X . 4. This can also be useful to gain insights into the optimal solution. (More on this later.)
Background image of page 8
Pros for Solving the Dual Review on Duality Pros and Cons Pros Cons Favorable Structures April 25, 2006 IE447 – 4 1. The dual is concave. By contrast, the primal may not be convex. 2. The dual may have smaller dimension, and has simpler constraints. 3. If there is no duality gap and the dual is solved exactly for a Lagrange multiplier μ , all optimal primal solutions can be obtained by minimizing the Lagrangian L ( x, μ ) over x X .
Background image of page 9

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

View Full DocumentRight Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 02/29/2008 for the course IE 417 taught by Professor Linderoth during the Spring '08 term at Lehigh University .

Page1 / 33

lecture20 - Duality Karush-Kuhn-Tucker and Lagrangian...

This preview shows document pages 1 - 10. Sign up to view the full document.

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