1
$\begingroup$

Theorem: Let $T:H\rightarrow K$ be a bounded linear operator between Hilbert spaces. Then there exists a unique bounded linear operator $T^* :K\rightarrow H$ such that $$\tag{1} \forall h\in H,k\in K :\langle Th,k\rangle=\langle h,T^*k\rangle. $$

The following proof of existence was given by by lecturer:

Suppose there is another linear operator, $S :K\rightarrow H$ such that $$\tag{2} \forall h\in H,k\in K :\langle Th,k\rangle=\langle h,Sk\rangle. $$ Then $$\tag{3} \left\langle h, T^{*} k-S k\right\rangle=\left\langle h, T^{*} k\right\rangle-\langle h, S k\rangle=\langle T h, k\rangle-\langle T h, k\rangle=0 . $$ Thus $T^{*} k=S k$ for all $k\in K$, and therefore $T^*=S$.

What confuses me is that in eq. $(2)$ he defined $S$ such that $\langle h,Sk\rangle=\langle h,T^*k\rangle$ for all $k\in K$, and I don't see why eq. $(3)$ makes it more obvious that $T^*=S$? What exactly is eq. $(3)$ showing that is not obvious from eq. $(1)$ and $(2)$?

$\endgroup$
1

1 Answer 1

1
$\begingroup$

This is using the fact that if a vector $v\in H$ satisfies $\langle h,v\rangle=0$ for all $h\in H$, then $v=0$. (To prove this, just take $h=v$ and use the positive-definiteness of the inner product.) Applying this with $v=T^*k-Sk$, equation (3) tells you that $T^*k-Sk=0$ so $T^*k=Sk$.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .