I'm trying to understand the proof of the following statement:
Statement: Let $X$ be a banach spase. We denote by $B(X,X)$ the set of all bounded linear operators from $X$ to $X$. Then the set of all invertible maps in $B(X,X)$ is open in $B(X,X)$.
Proof: Take invertible $T \in B(X,X)$. When $||S|| < ||T^{-1}||^{-1}, ||T^{-1}S||\leq ||T^{-1}|| \cdot||S|| < 1$.
By a proven Lemma we can say that $I+T^{-1}S$ is invertible. Then $T \cdot (I+T^{-1}S)$ is also invertible.
Question: I do not see how this proves the statement. I would have started in the following way:
Let $\Omega$ be the set of invertible functions s.t. $\Omega \subset B(X,X)$. My goal is to show that for any $T \in \Omega$ there exists a $\delta > 0$ such that $||S-T|| < \delta \implies S \in B(X,X)$.
From the question here: Open subsets of the space of linear operators I see that $||S-T|| < ||T^{-1}||^{-1}$ implies that $S$ is invertible. Using that I can conclude that $\Omega$ is open.
I don't get it how they proved this in the first proof. Following is not clear:
- I learned that if $||T^{-1}S|| < 1 \implies I -T^{-1}S$ invertible. Is the same true for $I +T^{-1}S$?
- The multiplication of two invertible operators is invertible. I get that. But how does the fact that $T \cdot (I+T^{-1}S)$ is invertible allow me to conclude the statement?