The spectral theorem guarantees, that when we have a symmetric matrix $A$, we can always decompose is into a multiplication of $PDP^T$, where:
- $P$ is an [[Orthogonal Matrix]] (square matrix with orthonormal row/column vectors)
- $D$ is a [[Matrix Diagonalization#Diagonal Matrix|Diagonal Matrix]].
$ A=PDP^T $
It turns out that the row or column vectors of $P$ are the [[Eigenvectors]] of $A$, and the diagonal elements of $D$ are the eigenvalues of $A$. We can demonstrate this linkage for by multiplying both sides by $v_1$.
$ Av_1=(PDP^T)v_1 $
![[spectral-theorem.jpeg|center|600]]
The matrix multiplications lead to the following, which is the definition of eigenvectors and eigenvalues of a matrix $A$.
$ Av_1=v_1\lambda_1 $