Is the Matrix Positive Semidefinite Given the Norm Condition?

  • Thread starter Thread starter brian_m.
  • Start date Start date
  • Tags Tags
    Matrix Positive
brian_m.
Messages
5
Reaction score
0
Hello.



Homework Statement



Let x \in \mathbb R^n and t \in \mathbb R.


Prove the following equivalence:
\left \| x \right \|_2 \leq t \ \ \Leftrightarrow \ \ \begin{pmatrix} t \cdot I_n & x \\ x^T & t \end{pmatrix} \text{is positive semidefinite }

Homework Equations



\left \| x \right \|_2 = \sqrt{x_1^2+ ... + x_n^2} is the euclidean norm and I_n the identity matrix of dimension n.


The Attempt at a Solution



I know that a matrix is positive semidefinite if and only if all eigenvalues of the matrix are \geq 0.
My problem is to calculate the eigenvalues of the given matrix.

Thank your for your help in advance!

Bye,
Brian
 
Physics news on Phys.org
yeah so I would probably start by trying to find the characteristic equation of the matrix

as they're only 2 non-zero rows in the first column, hopefully it shoudl simplify a fair bit
 
brian_m. said:
Hello.



Homework Statement



Let x \in \mathbb R^n and t \in \mathbb R.


Prove the following equivalence:
\left \| x \right \|_2 \leq t \ \ \Leftrightarrow \ \ \begin{pmatrix} t \cdot I_n & x \\ x^T & t \end{pmatrix} \text{is positive semidefinite }

Homework Equations



\left \| x \right \|_2 = \sqrt{x_1^2+ ... + x_n^2} is the euclidean norm and I_n the identity matrix of dimension n.


The Attempt at a Solution



I know that a matrix is positive semidefinite if and only if all eigenvalues of the matrix are \geq 0.
My problem is to calculate the eigenvalues of the given matrix.

Thank your for your help in advance!

Bye,
Brian

It is much easier if you forget about eigenvalues and look directly at the _defintion_ of psd (positive semi-definite). Your matrix A = \begin{pmatrix} t \cdot I_n & x \\ x^T & t \end{pmatrix} is psd if, for all Y \in \mathbb R^{n+1} we have
Y^T A Y \geq 0. Letting
Y = \begin{pmatrix} y \\ z \end{pmatrix}, \; y \in \mathbb{R}^n, \; z \in \mathbb{R}, the quadratic form Q(y,z) = Y^T A Y is easily computed. For alll y \in \mathbb{R}^n we need Q \geq 0, so considered as an optimization problem in z we can derive a simple necessary and sufficient condition involving ||x|| \text{ and } t.
 
Thanks for your help.

Now I have calculated Y^T A Y. It is:

Y^T A Y = \begin{pmatrix}<br /> y &amp; z \end{pmatrix} \begin{pmatrix} t \cdot I_n &amp; x \\ x^T &amp; t \end{pmatrix} \begin{pmatrix}<br /> y\\z \end{pmatrix} = \begin{pmatrix}<br /> y_1t+zx_1 &amp; \cdots &amp; y_nt+zx_n &amp; y_1x_1+...+y_nx_n+zt <br /> \end{pmatrix}\begin{pmatrix}<br /> y_1\\\vdots <br /> \\ y_n<br /> \\ z <br /> \end{pmatrix}= \\<br /> = \sum_{i=1}^n y_i^2 t + 2z \sum_{i=1}^n y_i x_i.

So I have to find out for which z the inequality t \cdot \sum_{i=1}^n y_i^2 + 2z \cdot \sum_{i=1}^n y_i x_i \geq 0 holds?

I don't know how to find out the solution of the inequality. Please can you help me again?

Thank you in adavance!

Bye,

Brian
 
brian_m. said:
Thanks for your help.

Now I have calculated Y^T A Y. It is:

Y^T A Y = \begin{pmatrix}<br /> y &amp; z \end{pmatrix} \begin{pmatrix} t \cdot I_n &amp; x \\ x^T &amp; t \end{pmatrix} \begin{pmatrix}<br /> y\\z \end{pmatrix} = \begin{pmatrix}<br /> y_1t+zx_1 &amp; \cdots &amp; y_nt+zx_n &amp; y_1x_1+...+y_nx_n+zt <br /> \end{pmatrix}\begin{pmatrix}<br /> y_1\\\vdots <br /> \\ y_n<br /> \\ z <br /> \end{pmatrix}= \\<br /> = \sum_{i=1}^n y_i^2 t + 2z \sum_{i=1}^n y_i x_i.

So I have to find out for which z the inequality t \cdot \sum_{i=1}^n y_i^2 + 2z \cdot \sum_{i=1}^n y_i x_i \geq 0 holds?

I don't know how to find out the solution of the inequality. Please can you help me again?

Thank you in adavance!

Bye,

Brian

Try again. I get Q(y,z) = Y^T A Y = t ||y||^2 + 2 &lt;x,y&gt; z + t z^2, where &lt;.,.&gt; denotes the inner product and ||\cdot || the usual norm. For any y, Q(y,z) must be ≥ 0, which means that as a function of z it cannot have two distinct roots.

RGV
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top