Hilbert Space Adjoint Operators
📂Hilbert SpaceHilbert Space Adjoint Operators
Buildup
Let us assume that we are given bounded linear operators T:K→H in Hilbert spaces (H,⟨⋅,⋅⟩H) and (K,⟨⋅,⋅⟩K). Then, for any fixed element w∈H, the following defined Φ:K→C becomes a linear functional Φ∈K∗.
Φv:=⟨Tv,w⟩H
According to the Riesz Representation Theorem, in the Hilbert space K, there must uniquely exist an element T∗w∈K for all v∈K that satisfies the following.
Φv=⟨v,T∗w⟩K
Although we cannot know what T∗w∈K specifically is for the previously fixed element w∈H, T∗ can be seen as the operator T∗:H→K that maps w to T∗w. From this discussion, the following concept can be introduced.
Definition
Let H,K be a Hilbert space. For a bounded linear operator T:K→H, the T∗:H→K that satisfies the following is called the adjoint operator of T.
⟨Tv,w⟩H=⟨v,T∗w⟩K,∀v∈K
Explanation
Also known as dual operator. It is also denoted by T#.
The adjoint operator has properties such as:
- T∗ is linear and bounded.
- (T∗)∗=T
- ∥T∗∥=∥T∥
Meanwhile, when H=K, adjoint operators that have the following good properties are called by a more special name. Let H be a Hilbert space and T:H→H be linear and bounded.
- If T=T∗, then T is called self-adjoint.
- If TT∗=T∗T=I, then T is called unitary.
If T is self-adjoint, then for all v,w∈H,
⟨Tv,w⟩=⟨v,Tw⟩
If T is unitary, then for all v,w∈H,
⟨Tv,Tw⟩=⟨v,w⟩
From the definition, unitary T is reversible,
T−1=T∗
Especially, unitary operators have a very important property related to the normal orthogonal basis.