Open mapping theorem (functional analysis)

In functional analysis, the open mapping theorem, also known as the Banach–Schauder theorem (named after Stefan Banach and Juliusz Schauder), is a fundamental result which states that if a continuous linear operator between Banach spaces is surjective then it is an open map. More precisely, (Rudin 1973, Theorem 2.11):

Open mapping theorem for Banach spacesEdit

Open mapping theorem for Banach spaces — If X and Y are Banach spaces and A : XY is a surjective continuous linear operator, then A is an open map (i.e. if U is an open set in X, then A(U) is open in Y).

One proof uses Baire's category theorem, and completeness of both X and Y is essential to the theorem. The statement of the theorem is no longer true if either space is just assumed to be a normed space, but is true if X and Y are taken to be Fréchet spaces.

Proof —

Suppose A : XY is a surjective continuous linear operator. In order to prove that A is an open map, it is sufficient to show that A maps the open unit ball in X to a neighborhood of the origin of Y.

Let  . Then

 .

Since A is surjective:

 .

But Y is Banach so by Baire's category theorem

 .

That is, we have cY and r > 0 such that

 .

Let vV, then

 .

By continuity of addition and linearity, the difference rv satisfies

 ,

and by linearity again,

 

where we have set L=2k/r. It follows that for all yY and all 𝜀 > 0, there exists some xX such that

 .

Our next goal is to show that VA(2LU).

Let yV. By (1), there is some x1 with ||x1|| < L and ||yAx1|| < 1/2. Define a sequence (xn) inductively as follows. Assume:

 

Then by (1) we can pick xn+1 so that:

 

so (2) is satisfied for xn+1. Let

 .

From the first inequality in (2), {sn} is a Cauchy sequence, and since X is complete, sn converges to some xX. By (2), the sequence Asn tends to y, and so Ax = y by continuity of A. Also,

 .

This shows that y belongs to A(2LU), so VA(2LU) as claimed. Thus the image A(U) of the unit ball in X contains the open ball V/2L of Y. Hence, A(U) is a neighborhood of 0 in Y, and this concludes the proof.

Related resultsEdit

Theorem[1] — Let X and Y be Banach spaces, let BX and BY denote their open unit balls, and let T : XY be a bounded linear operator. If δ > 0 then among the following four statements we have   (with the same δ)

  1.   for all  ;
  2.  ;
  3. T(BX) ⊆ δ BY;
  4. Im T = Y (i.e. T is surjective).

Furthermore, if T is surjective then (1) holds for some δ > 0

ConsequencesEdit

The open mapping theorem has several important consequences:

  • If A : XY is a bijective continuous linear operator between the Banach spaces X and Y, then the inverse operator A−1 : YX is continuous as well (this is called the bounded inverse theorem).[2]
  • If A : XY is a linear operator between the Banach spaces X and Y, and if for every sequence (xn) in X with xn → 0 and Axny it follows that y = 0, then A is continuous (the closed graph theorem).[3]

GeneralizationsEdit

Local convexity of X  or Y  is not essential to the proof, but completeness is: the theorem remains true in the case when X and Y are F-spaces. Furthermore, the theorem can be combined with the Baire category theorem in the following manner:

Theorem ((Rudin 1991, Theorem 2.11)) — Let X be a F-space and Y a topological vector space. If A : XY is a continuous linear operator, then either A(X) is a meager set in Y, or A(X) = Y. In the latter case, A is an open mapping and Y is also an F-space.

Furthermore, in this latter case if N is the kernel of A, then there is a canonical factorization of A in the form

 

where X / N is the quotient space (also an F-space) of X by the closed subspace N. The quotient mapping XX / N is open, and the mapping α is an isomorphism of topological vector spaces.[4]

Open mapping theorem ([5]) — If A : XY is a surjective closed linear operator from an complete pseudometrizable TVS X into a topological vector space Y and if at least one of the following conditions is satisfied:

  1. Y is a Baire space, or
  2. X is locally convex and Y is a barrelled space,

either A(X) is a meager set in Y, or A(X) = Y. then A is an open mapping.

Open mapping theorem for continuous maps ([5]) — Let A : XY be a continuous linear operator from an complete pseudometrizable TVS X into a Hausdorff topological vector space Y. If Im A is nonmeager in Y then A : XY is a surjective open map and Y is a complete pseudometrizable TVS.

The open mapping theorem can also be stated as

Theorem[6] — Let X and Y be two F-spaces. Then every continuous linear map of X onto Y is a TVS homomorphism, where a linear map u : XY is a topological vector space (TVS) homomorphism if the induced map   is a TVS-isomorphism onto its image.

ConsequencesEdit

Theorem[7] — If A : XY is a continuous linear bijection from a complete Pseudometrizable topological vector space (TVS) onto a Hausdorff TVS that is a Baire space, then A : XY is a homeomorphism (and thus an isomorphism of TVSs).

Webbed spacesEdit

Webbed spaces are a class of topological vector spaces for which the open mapping theorem and the closed graph theorem hold.

See alsoEdit

ReferencesEdit

  1. ^ Rudin 1991, p. 100.
  2. ^ Rudin 1973, Corollary 2.12.
  3. ^ Rudin 1973, Theorem 2.15.
  4. ^ Dieudonné 1970, 12.16.8.
  5. ^ a b Narici & Beckenstein 2011, p. 468.
  6. ^ Trèves 2006, p. 170
  7. ^ Narici & Beckenstein 2011, p. 469.

BibliographyEdit

This article incorporates material from Proof of open mapping theorem on PlanetMath, which is licensed under the Creative Commons Attribution/Share-Alike License.