Projective tensor product
In functional analysis, an area of mathematics, the projective tensor product of two locally convex topological vector spaces is a natural topological vector space structure on their tensor product. Namely, given locally convex topological vector spaces and
, the projective topology, or π-topology, on
is the strongest topology which makes
a locally convex topological vector space such that the canonical map
(from
to
) is continuous. When equipped with this topology,
is denoted
and called the projective tensor product of
and
. It is a particular instance of a topological tensor product.
Definitions
[edit]Let and
be locally convex topological vector spaces. Their projective tensor product
is the unique locally convex topological vector space with underlying vector space
having the following universal property:[1]
- For any locally convex topological vector space
, if
is the canonical map from the vector space of bilinear maps
to the vector space of linear maps
, then the image of the restriction of
to the continuous bilinear maps is the space of continuous linear maps
.
When the topologies of and
are induced by seminorms, the topology of
is induced by seminorms constructed from those on
and
as follows. If
is a seminorm on
, and
is a seminorm on
, define their tensor product
to be the seminorm on
given by
for all
in
, where
is the balanced convex hull of the set
. The projective topology on
is generated by the collection of such tensor products of the seminorms on
and
.[2][1]
When
and
are normed spaces, this definition applied to the norms on
and
gives a norm, called the projective norm, on
which generates the projective topology.[3]
Properties
[edit]Throughout, all spaces are assumed to be locally convex. The symbol denotes the completion of the projective tensor product of
and
.
- If
and
are both Hausdorff then so is
;[3] if
and
are Fréchet spaces then
is barelled.[4]
- For any two continuous linear operators
and
, their tensor product (as linear maps)
is continuous.[5]
- In general, the projective tensor product does not respect subspaces (e.g. if
is a vector subspace of
then the TVS
has in general a coarser topology than the subspace topology inherited from
).[6]
- If
and
are complemented subspaces of
and
respectively, then
is a complemented vector subspace of
and the projective norm on
is equivalent to the projective norm on
restricted to the subspace
. Furthermore, if
and
are complemented by projections of norm 1, then
is complemented by a projection of norm 1.[6]
- Let
and
be vector subspaces of the Banach spaces
and
, respectively. Then
is a TVS-subspace of
if and only if every bounded bilinear form on
extends to a continuous bilinear form on
with the same norm.[7]
Completion
[edit]In general, the space is not complete, even if both
and
are complete (in fact, if
and
are both infinite-dimensional Banach spaces then
is necessarily not complete[8]). However,
can always be linearly embedded as a dense vector subspace of some complete locally convex TVS, which is generally denoted by
.
The continuous dual space of is the same as that of
, namely, the space of continuous bilinear forms
.[9]
Grothendieck's representation of elements in the completion
[edit]In a Hausdorff locally convex space a sequence
in
is absolutely convergent if
for every continuous seminorm
on
[10] We write
if the sequence of partial sums
converges to
in
[10]
The following fundamental result in the theory of topological tensor products is due to Alexander Grothendieck.[11]
Theorem—Let and
be metrizable locally convex TVSs and let
Then
is the sum of an absolutely convergent series
where
and
and
are null sequences in
and
respectively.
The next theorem shows that it is possible to make the representation of independent of the sequences
and
Theorem[12]—Let and
be Fréchet spaces and let
(resp.
) be a balanced open neighborhood of the origin in
(resp. in
). Let
be a compact subset of the convex balanced hull of
There exists a compact subset
of the unit ball in
and sequences
and
contained in
and
respectively, converging to the origin such that for every
there exists some
such that
Topology of bi-bounded convergence
[edit]Let and
denote the families of all bounded subsets of
and
respectively. Since the continuous dual space of
is the space of continuous bilinear forms
we can place on
the topology of uniform convergence on sets in
which is also called the topology of bi-bounded convergence. This topology is coarser than the strong topology on
, and in (Grothendieck 1955), Alexander Grothendieck was interested in when these two topologies were identical. This is equivalent to the problem: Given a bounded subset
do there exist bounded subsets
and
such that
is a subset of the closed convex hull of
?
Grothendieck proved that these topologies are equal when and
are both Banach spaces or both are DF-spaces (a class of spaces introduced by Grothendieck[13]). They are also equal when both spaces are Fréchet with one of them being nuclear.[9]
Strong dual and bidual
[edit]Let be a locally convex topological vector space and let
be its continuous dual space. Alexander Grothendieck characterized the strong dual and bidual for certain situations:
Theorem[14] (Grothendieck)—Let and
be locally convex topological vector spaces with
nuclear. Assume that both
and
are Fréchet spaces, or else that they are both DF-spaces. Then, denoting strong dual spaces with a subscripted
:
- The strong dual of
can be identified with
;
- The bidual of
can be identified with
;
- If
is reflexive then
(and hence
) is a reflexive space;
- Every separately continuous bilinear form on
is continuous;
- Let
be the space of bounded linear maps from
to
. Then, its strong dual can be identified with
so in particular if
is reflexive then so is
Examples
[edit]- For
a measure space, let
be the real Lebesgue space
; let
be a real Banach space. Let
be the completion of the space of simple functions
, modulo the subspace of functions
whose pointwise norms, considered as functions
, have integral
with respect to
. Then
is isometrically isomorphic to
.[15]
See also
[edit]- Inductive tensor product
- Injective tensor product
- Tensor product of Hilbert spaces – Tensor product space endowed with a special inner product
Citations
[edit]- 1 2 Trèves 2006, p. 438.
- ↑ Trèves 2006, p. 435.
- 1 2 Trèves 2006, p. 437.
- ↑ Trèves 2006, p. 445.
- ↑ Trèves 2006, p. 439.
- 1 2 Ryan 2002, p. 18.
- ↑ Ryan 2002, p. 24.
- ↑ Ryan 2002, p. 43.
- 1 2 Schaefer & Wolff 1999, p. 173.
- 1 2 Schaefer & Wolff 1999, p. 120.
- ↑ Schaefer & Wolff 1999, p. 94.
- ↑ Trèves 2006, pp. 459–460.
- ↑ Schaefer & Wolff 1999, p. 154.
- ↑ Schaefer & Wolff 1999, pp. 175–176.
- ↑ Schaefer & Wolff 1999, p. 95.
References
[edit]- Ryan, Raymond (2002). Introduction to tensor products of Banach spaces. London New York: Springer. ISBN 1-85233-437-1. OCLC 48092184.
- Schaefer, Helmut H.; Wolff, Manfred P. (1999). Topological Vector Spaces. GTM. Vol. 8 (Second ed.). New York, NY: Springer New York Imprint Springer. ISBN 978-1-4612-7155-0. OCLC 840278135.
- Trèves, François (2006) [1967]. Topological Vector Spaces, Distributions and Kernels. Mineola, N.Y.: Dover Publications. ISBN 978-0-486-45352-1. OCLC 853623322.
Further reading
[edit]- Diestel, Joe (2008). The metric theory of tensor products : Grothendieck's résumé revisited. Providence, R.I: American Mathematical Society. ISBN 978-0-8218-4440-3. OCLC 185095773.
- Grothendieck, Alexander (1955). "Produits Tensoriels Topologiques et Espaces Nucléaires" [Topological Tensor Products and Nuclear Spaces]. Memoirs of the American Mathematical Society Series (in French). 16. Providence: American Mathematical Society. MR 0075539. OCLC 9308061.
- Grothendieck, Grothendieck (1966). Produits tensoriels topologiques et espaces nucléaires (in French). Providence: American Mathematical Society. ISBN 0-8218-1216-5. OCLC 1315788.
- Pietsch, Albrecht (1972). Nuclear locally convex spaces. Berlin, New York: Springer-Verlag. ISBN 0-387-05644-0. OCLC 539541.
- Wong (1979). Schwartz spaces, nuclear spaces, and tensor products. Berlin New York: Springer-Verlag. ISBN 3-540-09513-6. OCLC 5126158.