Menu
QuantML.orgQuantML.org
  • Linear Algebra
  • Null Space
  • Echelon Form
  • Solve
    \(A\vec{x}=0\)
  • RREF

    Previously we saw how to get echelon form, pivot/free column vectors of a matrix say

    \(A\)
    also saw how to solve
    \(A\vec{x}=0\)
    .
    Now let's push it little further and see the reduce row echelon form of our matrix
    \(A\)
    .

    Reduced Row Echelon Form

    In reduced row echelon form we use elimination and make all elements
    \(0\)
    above and below pivot elements.
    We saw the echelon form of
    \(A=\begin{bmatrix} 1 & 2 & 2 & 2\\ 2 & 4 & 6 & 8\\ 3 & 6 & 8 & 10\\ \end{bmatrix}\)
    is say
    \(U\)

    \(U=\begin{bmatrix} \fbox{1} & 2 & 2 & 2\\ 0 & 0 & \fbox{2} & 4\\ 0 & 0 & 0 & 0\\ \end{bmatrix}\)

    Now we make element above pivot
    \(=0\)

    So let's use elimination to do it.
    \(U=\begin{bmatrix} \fbox{1} & 2 & 2 & 2\\ 0 & 0 & \fbox{2} & 4\\ 0 & 0 & 0 & 0\\ \end{bmatrix}\)

    \(r_1\leftarrow r_1-r_2\)

    \(U'=\begin{bmatrix} \fbox{1} & 2 & 0 & -2\\ 0 & 0 & \fbox{2} & 4\\ 0 & 0 & 0 & 0\\ \end{bmatrix}\)

    Now make pivot elements
    \(=1\)

    \(r_2\leftarrow r_2/2\)

    \(U'=\begin{bmatrix} \fbox{1} & 2 & 0 & -2\\ 0 & 0 & \fbox{1} & 2\\ 0 & 0 & 0 & 0\\ \end{bmatrix}\)

    Say this matrix is
    \(R\)
    .
    \(R=\begin{bmatrix} \color{red}{\fbox{1}} & \color{blue}{2} & \color{red}{0} & \color{blue}{-2} \\ \color{red}{0} & \color{blue}{0} & \color{red}{\fbox{1}} & \color{blue}{2} \\ 0 & 0 & 0 & 0 \\ \end{bmatrix}\)

    Here you can see that there is a
    \(2\times 2\)
    Identity matrix shown in
    \(\color{red}{\text{red}}\)

    and one more
    \(2\times 2\)
    matrix shown in
    \(\color{blue}{\text{blue}}\)

    So
    \(R\)
    is something like:
    \(R'=\begin{bmatrix} \color{red}{\text{I}} & \color{blue}{\text{F}} \\ 0 & 0 \\ \end{bmatrix}\)

    and here
    \(\color{red}{ I=\begin{bmatrix} 1&0\\0&1\\ \end{bmatrix} }\)
    and
    \(\color{blue}{ F=\begin{bmatrix} 2&-2\\0&2\\ \end{bmatrix} }\)

    Here you can see that there is some reordering of columns but it's fine.
    Because we are striving for
    \(\vec{x}\)
    that solves
    \(R\vec{x}=0\)

    \(R\vec{x}=\begin{bmatrix} \color{red}{\fbox{1}} & \color{blue}{2} & \color{red}{0} & \color{blue}{-2} \\ \color{red}{0} & \color{blue}{0} & \color{red}{\fbox{1}} & \color{blue}{2} \\ 0 & 0 & 0 & 0 \\ \end{bmatrix} \begin{bmatrix} \color{red}{x_1}\\\color{blue}{x_2}\\\color{red}{x_3}\\\color{blue}{x_4}\\ \end{bmatrix} =0\)

    We can also write it as:
    \(R'\vec{x}'=\begin{bmatrix} \color{red}{\fbox{1}} & \color{red}{0} & \color{blue}{2} & \color{blue}{-2} \\ \color{red}{0} & \color{red}{\fbox{1}} & \color{blue}{0} & \color{blue}{2} \\ 0 & 0 & 0 & 0 \\ \end{bmatrix} \begin{bmatrix} \color{red}{x_1}\\\color{red}{x_3}\\\color{blue}{x_2}\\\color{blue}{x_4} \\ \end{bmatrix} =0\)

    Now you can see
    \(R\)
    as:
    \(R'=\begin{bmatrix} \color{red}{\text{I}} & \color{blue}{\text{F}} \\ 0 & 0 \\ \end{bmatrix}\)
    Remember that number of special solution is
    \(\color{blue}{2}\)
    because we have 2 free column vectors.
    So we know that there are
    \(\color{blue}{2}\)
    special solution possible.
    Say that special solutions are:
    \(\vec{x}^{(1)}=\begin{bmatrix} \color{red}{x^{(1)}_1}\\\color{blue}{x^{(1)}_2}\\\color{red}{x^{(1)}_3}\\\color{blue}{x^{(1)}_4}\\ \end{bmatrix}\)
    and
    \(\vec{x}^{(2)}=\begin{bmatrix} \color{red}{x^{(2)}_1}\\\color{blue}{x^{(2)}_2}\\\color{red}{x^{(2)}_3}\\\color{blue}{x^{(2)}_4}\\ \end{bmatrix}\)

    Now rearrange then
    \(\vec{x}'^{(1)}=\begin{bmatrix} \color{red}{x^{(1)}_1}\\\color{red}{x^{(1)}_3}\\\color{blue}{x^{(1)}_2}\\\color{blue}{x^{(1)}_4}\\ \end{bmatrix}\)
    and
    \(\vec{x}'^{(2)}=\begin{bmatrix} \color{red}{x^{(2)}_1}\\\color{red}{x^{(2)}_3}\\\color{blue}{x^{(2)}_2}\\\color{blue}{x^{(2)}_4}\\ \end{bmatrix}\)

    Now stack then in a matrix say (
    \(N\)
    )
    \(N=\begin{bmatrix} \color{red}{x^{(1)}_1} & \color{red}{x^{(2)}_1} \\ \color{red}{x^{(1)}_3} & \color{red}{x^{(2)}_3} \\ \color{blue}{x^{(1)}_2} & \color{blue}{x^{(2)}_2} \\ \color{blue}{x^{(1)}_4} & \color{blue}{x^{(2)}_4} \\ \end{bmatrix};\quad\)
    and say
    \(x_{\text{pivot}}=\begin{bmatrix} \color{red}{x^{(1)}_1} & \color{red}{x^{(2)}_1} \\ \color{red}{x^{(1)}_3} & \color{red}{x^{(2)}_3} \\ \end{bmatrix}\)
    \(x_{\text{free}} =\begin{bmatrix} \color{blue}{x^{(1)}_2} & \color{blue}{x^{(2)}_2} \\ \color{blue}{x^{(1)}_4} & \color{blue}{x^{(2)}_4} \\ \end{bmatrix}\)

    And we are free to choose free variables so let
    \(x_{\text{free}} =\begin{bmatrix} \color{blue}{1} & \color{blue}{0} \\ \color{blue}{0} & \color{blue}{1} \\ \end{bmatrix}\)

    Now
    \(N=\begin{bmatrix} \color{red}{x_\text{pivot}}\\\color{blue}{I_{2\times2}} \end{bmatrix}\)

    We want
    \(R\vec{x}^{(1)}=0\)
    and
    \(R\vec{x}^{(2)}=0\)

    \(\Rightarrow R'\vec{x}'^{(1)}=0\)
    and
    \(R'\vec{x}'^{(2)}=0\)

    \(\Rightarrow R'N=0\)

    So
    \(R'N=\begin{bmatrix} \color{red}{I_{2\times2}} & \color{blue}{\text{F}} \\ 0 & 0 \\ \end{bmatrix} \begin{bmatrix} \color{red}{x_\text{pivot}}\\\color{blue}{I_{2\times2}} \end{bmatrix} = 0 \)

    \(\Rightarrow \color{red}{x_\text{pivot}} \color{red}{I_{2\times2}} + \color{blue}{\text{F}} \color{blue}{I_{2\times2}} =0 \)

    \(\Rightarrow \color{red}{x_\text{pivot}} = - \color{blue}{\text{F}}\)

    So
    \(N=\begin{bmatrix} - \color{blue}{\text{F}} \\ \color{blue}{I_{2\times2}}\\ \end{bmatrix} \)

    And we know that
    \(\color{blue}{ F=\begin{bmatrix} 2&-2\\0&2\\ \end{bmatrix} }\)

    So
    \(N=\begin{bmatrix} \color{red}{-2} & \color{red}{2} \\ \color{red}{0} & \color{red}{-2} \\ \color{blue}{1} & \color{blue}{0} \\ \color{blue}{0} & \color{blue}{1} \\ \end{bmatrix};\quad\)

    \(\Rightarrow\)
    \(\vec{x}'^{(1)}=\begin{bmatrix} \color{red}{x^{(1)}_1}\\\color{red}{x^{(1)}_3}\\\color{blue}{x^{(1)}_2}\\\color{blue}{x^{(1)}_4}\\ \end{bmatrix} = \begin{bmatrix} \color{red}{-2}\\\color{red}{0}\\\color{blue}{1}\\\color{blue}{0}\\ \end{bmatrix}\)
    and
    \(\vec{x}'^{(2)}=\begin{bmatrix} \color{red}{x^{(2)}_1}\\\color{red}{x^{(2)}_3}\\\color{blue}{x^{(2)}_2}\\\color{blue}{x^{(2)}_4}\\ \end{bmatrix} = \begin{bmatrix} \color{red}{2}\\\color{red}{-2}\\\color{blue}{0}\\\color{blue}{1}\\ \end{bmatrix}\)

    \(\Rightarrow\)
    \(\vec{x}^{(1)}=\begin{bmatrix} \color{red}{x^{(1)}_1}\\\color{blue}{x^{(1)}_2}\\\color{red}{x^{(1)}_3}\\\color{blue}{x^{(1)}_4}\\ \end{bmatrix} = \begin{bmatrix} \color{red}{-2}\\\color{red}{1}\\\color{blue}{0}\\\color{blue}{0}\\ \end{bmatrix}\)
    and
    \(\vec{x}^{(2)}=\begin{bmatrix} \color{red}{x^{(2)}_1}\\\color{blue}{x^{(2)}_2}\\\color{red}{x^{(2)}_3}\\\color{blue}{x^{(2)}_4}\\ \end{bmatrix} = \begin{bmatrix} \color{red}{2}\\\color{red}{0}\\\color{blue}{-2}\\\color{blue}{1}\\ \end{bmatrix}\)


    So our Null space is space of all
    \(\vec{x}\)
    that are linear combinations of
    \(\begin{bmatrix} -2\\1\\0\\0\\ \end{bmatrix} \)
    and
    \(\begin{bmatrix} 2\\0\\-2\\1\\ \end{bmatrix}\)

    \(x=\alpha \begin{bmatrix} -2\\1\\0\\0\\ \end{bmatrix} + \beta \begin{bmatrix} 2\\0\\-2\\1\\ \end{bmatrix};\quad\)
    \(\alpha,\beta\in\mathbb{R}\)