Let A be a 4x4 matrix with eigenvalues $\lambda$ = 2,3 and eigenspaces
$E_{\lambda=2} = \operatorname{span} \left\{ {\begin{bmatrix} 1\\0\\0\\1\end{bmatrix}, \begin{bmatrix} 1\\0\\1\\0\end{bmatrix}}\right\}$ and $E_{\lambda=3} = \operatorname{span} \left\{ {\begin{bmatrix} 0\\1\\0\\1\end{bmatrix}, \begin{bmatrix} 0\\0\\1\\1\end{bmatrix}}\right\}$
(A) Can you calculate the algebraic multiplicities of $\lambda$ = 2 and $\lambda$ = 3? If yes, do it.
(B) Can you calculate A? If yes, how?
I know how to calculate the algebraic multiplicities given the matrix A, but I do not know where to start when I am only given the eigenvalues and eigenspaces. My guess as to how to find A is to use the theorem A $\vec x$ = $\lambda$$\vec x$
For example, later in the problem I am given $\vec x = \begin{bmatrix} 3\\1\\4\\6\end{bmatrix} $
Solving A $\vec x$ = $\lambda$ $\vec x$, I get $A\begin{bmatrix} 3\\1\\4\\6\end{bmatrix} = \begin{bmatrix} 6\\3\\11\\16\end{bmatrix}$
How do I find A from here?
Additionally, I know that I cannot recover A from the eigenpairs if A is deficient, so I know I need to solve (a) first in order to solve (b). However, I have no idea how to find the algebraic multiplicities of A without first knowing A and finding the characteristic polynomial.
A Long Hint: (thanks for your compliance and your patience!)
Recall that the geometric multiplicities are always at most the algebraic multiplicities; in other words $A(\lambda)\geq G(\lambda)$. By the information we already have, we know that $A(2)\geq 2$ and $A(3)\geq 2$.
Moreover, some general theory: we know that the characteristic polynomial has degree $n$, and by the Fundamental Theorem of Algebra every degree-$n$ real polynomial has $n$ complex roots. Therefore, $\sum_\lambda A(\lambda)=n$.
Applying the general theory to our situation, we have $A(2)+A(3)=4$.
Can you see how to proceed?