Bourbaki

Definition 1

Tag 080Y, permanent. Cite it as [Bourbaki, Tag 080Y].

Topological Vector Spaces, chapter V, § 4. Some classes of operators in hilbertian spaces

— Let E and F be two hilbertian spaces. For every continuous linear mapping u:E→Fu : E \to F, the continuous linear mapping from F into E defined by formula (1) is called the adjoint of u and is denoted by u∗u^*.

We have

(u+v)∗=u∗+v∗(λu)∗=λ‾u∗(u∗)∗=u(1E)∗=1E(wu)∗=u∗w∗; \begin{align*} (u + v)^* &= u^* + v^* \\ (\lambda u)^* &= \overline{\lambda} u^* \\ (u^*)^* &= u \\ (1_E)^* &= 1_E \\ (wu)^* &= u^* w^* ; \end{align*}

in all these formulas, uu and vv belong to L(E;F)\mathscr{L}(E; F), λ\lambda is in K, and ww in L(F;G)\mathscr{L}(F; G) where G is a hilbertian space. Formulas (5) and (6) mean that u↦u∗u \mapsto u^* is semi-linear. Formula (8) is obvious. To prove (7), we take the conjugate of the two members of (1), which gives ⟨u∗(y)∣x⟩=⟨y∣u(x)⟩\langle u^*(y)|x\rangle = \langle y|u(x)\rangle, and this proves that uu is the adjoint of u∗u^*. Finally, with the notations of (9), we have, for all z∈Gz \in G

⟨w(u(x))∣z⟩=⟨u(x)∣w∗(z)⟩=⟨x∣u∗(w∗(z))⟩, \langle w(u(x))|z\rangle = \langle u(x)|w^*(z)\rangle = \langle x|u^*(w^*(z))\rangle ,

hence u∗w∗u^*w^* is the adjoint of wuwu.

Let u:E→Fu : E \to F be a bijective and continuous linear mapping; then it is also bicontinuous (I, p. 19, cor. 1). From (8) and (9) we immediately deduce that u∗u^* is bijective and bicontinuous and that

(10)

(u−1)∗=(u∗)−1. (u^{-1})^* = (u^*)^{-1} .

Bourbaki puts the proof directly after the statement, so what is above is the statement and its proof together.

Cited by

Nothing.

Cites

Nothing.

Transcribed from the printed volume. Not translated.