Line 1: Line 1:
 
= ECE QE AC-3 August 2011 Solusion  =
 
= ECE QE AC-3 August 2011 Solusion  =
  
==== <font face="serif"><math>\text{1.} \left( \text{20 pts} \right) \text{Consider the optimization problem,}</math><br></font>  ====
+
==== <span class="texhtml">1. (20 pts) Consider the optimization problem,</span><br>  ====
  
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<math>\text{maximize} -x_{1}^{2}+x_{1}-x_{2}-x_{1}x_{2}</math>  
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<math>\text{maximize} -x_{1}^{2}+x_{1}-x_{2}-x_{1}x_{2}</math>  
Line 7: Line 7:
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<math>\text{subject to  }  x_{1}\geq0, x_{2}\geq0</math>  
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;<math>\text{subject to  }  x_{1}\geq0, x_{2}\geq0</math>  
  
<math>\left( \text{i} \right) \text{Characterize feasible directions at the point } x^{*}=\left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right]</math>  
+
<span class="texhtml">(i) Characterize feasible directions at the point</span><math>x^{*}=\left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right]</math>  
  
 
===== <math>\color{blue}\text{Solution 1:}</math>  =====
 
===== <math>\color{blue}\text{Solution 1:}</math>  =====
Line 13: Line 13:
 
<math>\text{We need to find a direction }d\text{, such that } \exists\alpha_{0}>0,</math><br>  
 
<math>\text{We need to find a direction }d\text{, such that } \exists\alpha_{0}>0,</math><br>  
  
&nbsp;<math>d\in\Re_{2}, d\neq0 \text{ is a feasible direction at } \\ x^{*} \text{, if } \exists\alpha_{0} \\ \text{that } \left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right] + \alpha\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right] \in\Omega \text{ for all } 0\leq\alpha\leq\alpha_{0}</math>&nbsp;
+
<math>\color{blue}\text{Solution 2:}</math>
 +
 
 +
<math>d\in\Re_{2}, d\neq0 \text{ is a feasible direction at } x^{*} \text{, if}</math>&nbsp;<math>\exists\alpha_{0}</math>&nbsp;  
  
 
<math>\text{that } \left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right] + \alpha\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right] \in\Omega \text{ for all } 0\leq\alpha\leq\alpha_{0}</math>&nbsp;<br>  
 
<math>\text{that } \left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right] + \alpha\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right] \in\Omega \text{ for all } 0\leq\alpha\leq\alpha_{0}</math>&nbsp;<br>  
Line 22: Line 24:
 
\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right], d_{1}\in\Re^{2}, d_{2}\neq0</math><br>  
 
\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right], d_{1}\in\Re^{2}, d_{2}\neq0</math><br>  
  
===== <math>\left( \text{ii} \right) \text{Write down the second-order necessary condition for } x^{*} \text{. Does the point } x^{*} \text{ satisfy this condition?}</math><br> =====
+
===== (ii) Write down the second-order necessary condition for . Does the point satisfy this condition? =====

Revision as of 21:32, 21 June 2012

ECE QE AC-3 August 2011 Solusion

1. (20 pts) Consider the optimization problem,

               $ \text{maximize} -x_{1}^{2}+x_{1}-x_{2}-x_{1}x_{2} $

               $ \text{subject to } x_{1}\geq0, x_{2}\geq0 $

(i) Characterize feasible directions at the point$ x^{*}=\left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right] $

$ \color{blue}\text{Solution 1:} $

$ \text{We need to find a direction }d\text{, such that } \exists\alpha_{0}>0, $

$ \color{blue}\text{Solution 2:} $

$ d\in\Re_{2}, d\neq0 \text{ is a feasible direction at } x^{*} \text{, if} $ $ \exists\alpha_{0} $ 

$ \text{that } \left[ \begin{array}{c} \frac{1}{2} \\ 0 \end{array} \right] + \alpha\left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right] \in\Omega \text{ for all } 0\leq\alpha\leq\alpha_{0} $ 

$ \because \begin{Bmatrix}x\in\Omega: x_{1}\geq0, x_{2}\geq0\end{Bmatrix} $


$ \therefore d= \left[ \begin{array}{c} d_{1} \\ d_{2} \end{array} \right], d_{1}\in\Re^{2}, d_{2}\neq0 $

(ii) Write down the second-order necessary condition for . Does the point satisfy this condition?

Alumni Liaison

Abstract algebra continues the conceptual developments of linear algebra, on an even grander scale.

Dr. Paul Garrett