In mathematics, particularly functional analysis, spaces of linear maps between two vector spaces can be endowed with a variety of topologies. Studying space of linear maps and these topologies can give insight into the spaces themselves.
The article operator topologies discusses topologies on spaces of linear maps between normed spaces, whereas this article discusses topologies on such spaces in the more general setting of topological vector spaces (TVSs).
Throughout, the following is assumed:
- is any non-empty set and is a non-empty collection of subsets of directed by subset inclusion (i.e. for any there exists some such that ).
- is a topological vector space (not necessarily Hausdorff or locally convex).
- is a basis of neighborhoods of 0 in
- is a vector subspace of [note 1] which denotes the set of all -valued functions with domain
𝒢-topology
The following sets will constitute the basic open subsets of topologies on spaces of linear maps.
For any subsets and let
The family
forms a neighborhood basis[1]
at the origin for a unique translation-invariant topology on where this topology is not necessarily a vector topology (that is, it might not make into a TVS).
This topology does not depend on the neighborhood basis that was chosen and it is known as the topology of uniform convergence on the sets in or as the -topology.
However, this name is frequently changed according to the types of sets that make up (e.g. the "topology of uniform convergence on compact sets" or the "topology of compact convergence", see the footnote for more details[3]).
A subset of is said to be fundamental with respect to if each is a subset of some element in
In this case, the collection can be replaced by without changing the topology on
One may also replace with the collection of all subsets of all finite unions of elements of without changing the resulting -topology on
Call a subset of -bounded if is a bounded subset of for every
Properties
Properties of the basic open sets will now be described, so assume that and
Then is an absorbing subset of if and only if for all absorbs .
If is balanced (respectively, convex) then so is
The equality
always holds.
If is a scalar then so that in particular,
Moreover,
and similarly
For any subsets and any non-empty subsets
which implies:
- if then
- if then
- For any and subsets of if then
For any family of subsets of and any family of neighborhoods of the origin in
For any and be any entourage of (where is endowed with its canonical uniformity), let
Given the family of all sets as ranges over any fundamental system of entourages of forms a fundamental system of entourages for a uniform structure on called the uniformity of uniform converges on or simply the -convergence uniform structure.
The -convergence uniform structure is the least upper bound of all -convergence uniform structures as ranges over
Nets and uniform convergence
Let and let be a net in Then for any subset of say that converges uniformly to on if for every there exists some such that for every satisfying (or equivalently, for every ).
Throughout this section we will assume that and are topological vector spaces.
will be a non-empty collection of subsets of directed by inclusion.
will denote the vector space of all continuous linear maps from into If is given the -topology inherited from then this space with this topology is denoted by .
The continuous dual space of a topological vector space over the field (which we will assume to be real or complex numbers) is the vector space and is denoted by .
The -topology on is compatible with the vector space structure of if and only if for all and all the set is bounded in which we will assume to be the case for the rest of the article.
Note in particular that this is the case if consists of (von-Neumann) bounded subsets of
Assumptions on 𝒢
Assumptions that guarantee a vector topology
- ( is directed): will be a non-empty collection of subsets of directed by (subset) inclusion. That is, for any there exists such that .
The above assumption guarantees that the collection of sets forms a filter base.
The next assumption will guarantee that the sets are balanced.
Every TVS has a neighborhood basis at 0 consisting of balanced sets so this assumption isn't burdensome.
- ( are balanced): is a neighborhoods basis of the origin in that consists entirely of balanced sets.
The following assumption is very commonly made because it will guarantee that each set is absorbing in
- ( are bounded): is assumed to consist entirely of bounded subsets of
The next theorem gives ways in which can be modified without changing the resulting -topology on
Common assumptions
Some authors (e.g. Narici) require that satisfy the following condition, which implies, in particular, that is directed by subset inclusion:
- is assumed to be closed with respect to the formation of subsets of finite unions of sets in (i.e. every subset of every finite union of sets in belongs to ).
Some authors (e.g. Trèves ) require that be directed under subset inclusion and that it satisfy the following condition:
- If and is a scalar then there exists a such that
If is a bornology on which is often the case, then these axioms are satisfied.
If is a saturated family of bounded subsets of then these axioms are also satisfied.
Properties
Hausdorffness
A subset of a TVS whose linear span is a dense subset of is said to be a total subset of
If is a family of subsets of a TVS then is said to be total in if the linear span of is dense in
If is the vector subspace of consisting of all continuous linear maps that are bounded on every then the -topology on is Hausdorff if is Hausdorff and is total in
Completeness
For the following theorems, suppose that is a topological vector space and is a locally convex Hausdorff spaces and is a collection of bounded subsets of that covers is directed by subset inclusion, and satisfies the following condition: if and is a scalar then there exists a such that
- is complete if
- is locally convex and Hausdorff,
- is complete, and
- whenever is a linear map then restricted to every set is continuous implies that is continuous,
- If is a Mackey space then is complete if and only if both and are complete.
- If is barrelled then is Hausdorff and quasi-complete.
- Let and be TVSs with quasi-complete and assume that (1) is barreled, or else (2) is a Baire space and and are locally convex. If covers then every closed equicontinuous subset of is complete in and is quasi-complete.
- Let be a bornological space, a locally convex space, and a family of bounded subsets of such that the range of every null sequence in is contained in some If is quasi-complete (respectively, complete) then so is .
Boundedness
Let and be topological vector spaces and be a subset of
Then the following are equivalent:
- is bounded in ;
- For every is bounded in ;
- For every neighborhood of the origin in the set absorbs every
If is a collection of bounded subsets of whose union is total in then every equicontinuous subset of is bounded in the -topology.
Furthermore, if and are locally convex Hausdorff spaces then
- if is bounded in (that is, pointwise bounded or simply bounded) then it is bounded in the topology of uniform convergence on the convex, balanced, bounded, complete subsets of
- if is quasi-complete (meaning that closed and bounded subsets are complete), then the bounded subsets of are identical for all -topologies where is any family of bounded subsets of covering
Examples
The topology of pointwise convergence
By letting be the set of all finite subsets of will have the weak topology on or the topology of pointwise convergence or the topology of simple convergence and with this topology is denoted by .
Unfortunately, this topology is also sometimes called the strong operator topology, which may lead to ambiguity; for this reason, this article will avoid referring to this topology by this name.
A subset of is called simply bounded or weakly bounded if it is bounded in .
The weak-topology on has the following properties:
- If is separable (that is, it has a countable dense subset) and if is a metrizable topological vector space then every equicontinuous subset of is metrizable; if in addition is separable then so is
- So in particular, on every equicontinuous subset of the topology of pointwise convergence is metrizable.
- Let denote the space of all functions from into If is given the topology of pointwise convergence then space of all linear maps (continuous or not) into is closed in .
- In addition, is dense in the space of all linear maps (continuous or not) into
- Suppose and are locally convex. Any simply bounded subset of is bounded when has the topology of uniform convergence on convex, balanced, bounded, complete subsets of If in addition is quasi-complete then the families of bounded subsets of are identical for all -topologies on such that is a family of bounded sets covering
Equicontinuous subsets
- The weak-closure of an equicontinuous subset of is equicontinuous.
- If is locally convex, then the convex balanced hull of an equicontinuous subset of is equicontinuous.
- Let and be TVSs and assume that (1) is barreled, or else (2) is a Baire space and and are locally convex. Then every simply bounded subset of is equicontinuous.
- On an equicontinuous subset of the following topologies are identical: (1) topology of pointwise convergence on a total subset of ; (2) the topology of pointwise convergence; (3) the topology of precompact convergence.
Throughout, we assume that is a TVS.
List of polar topologies
Suppose that is a TVS whose bounded subsets are the same as its weakly bounded subsets.
Notation: If denotes a polar topology on then endowed with this topology will be denoted by or simply (e.g. for we would have so that and all denote with endowed with ).
More information > ...
> ("topology of uniform convergence on ...") |
Notation |
Name ("topology of...") |
Alternative name |
finite subsets of |
|
pointwise/simple convergence |
weak/weak* topology |
-compact disks |
|
|
Mackey topology |
-compact convex subsets |
|
compact convex convergence |
|
-compact subsets (or balanced -compact subsets) |
|
compact convergence |
|
-bounded subsets |
|
bounded convergence |
strong topology |
Close
We will let denote the space of separately continuous bilinear maps and denote the space of continuous bilinear maps, where and are topological vector space over the same field (either the real or complex numbers).
In an analogous way to how we placed a topology on we can place a topology on and .
Let (respectively, ) be a family of subsets of (respectively, ) containing at least one non-empty set.
Let denote the collection of all sets where
We can place on the -topology, and consequently on any of its subsets, in particular on and on .
This topology is known as the -topology or as the topology of uniform convergence on the products of .
However, as before, this topology is not necessarily compatible with the vector space structure of or of without the additional requirement that for all bilinear maps, in this space (that is, in or in ) and for all and the set is bounded in
If both and consist of bounded sets then this requirement is automatically satisfied if we are topologizing but this may not be the case if we are trying to topologize .
The -topology on will be compatible with the vector space structure of if both and consists of bounded sets and any of the following conditions hold:
- and are barrelled spaces and is locally convex.
- is a F-space, is metrizable, and is Hausdorff, in which case
- and are the strong duals of reflexive Fréchet spaces.
- is normed and and the strong duals of reflexive Fréchet spaces.
The ε-topology
Suppose that and are locally convex spaces and let and be the collections of equicontinuous subsets of and , respectively.
Then the -topology on will be a topological vector space topology.
This topology is called the ε-topology and with this topology it is denoted by or simply by
Part of the importance of this vector space and this topology is that it contains many subspace, such as which we denote by
When this subspace is given the subspace topology of it is denoted by
In the instance where is the field of these vector spaces, is a tensor product of and
In fact, if and are locally convex Hausdorff spaces then is vector space-isomorphic to which is in turn is equal to
These spaces have the following properties:
- If and are locally convex Hausdorff spaces then is complete if and only if both and are complete.
- If and are both normed (respectively, both Banach) then so is