topology (point-set topology, point-free topology)
see also differential topology, algebraic topology, functional analysis and topological homotopy theory
Basic concepts
fiber space, space attachment
Extra stuff, structure, properties
Kolmogorov space, Hausdorff space, regular space, normal space
sequentially compact, countably compact, locally compact, sigma-compact, paracompact, countably paracompact, strongly compact
Examples
Basic statements
closed subspaces of compact Hausdorff spaces are equivalently compact subspaces
open subspaces of compact Hausdorff spaces are locally compact
compact spaces equivalently have converging subnet of every net
continuous metric space valued function on compact metric space is uniformly continuous
paracompact Hausdorff spaces equivalently admit subordinate partitions of unity
injective proper maps to locally compact spaces are equivalently the closed embeddings
locally compact and second-countable spaces are sigma-compact
Theorems
Analysis Theorems
A function is continuous if, roughly speaking, is arbitrarily close to whenever is sufficiently close to . However, ‘close’ is relative, and may be much closer to than is to , even if both and are continuous. Nevertheless, given a family of functions, we may have that is arbitrarily close to for every function in the family at once whenever is sufficiently close to . In this case, the family of functions is equicontinuous.
Because we are considering the relative degree of closeness between potentially unrelated pairs of points, we need a uniform structure to define this concept. So let and be uniform spaces (although the concept should make sense in somewhat greater generality), and let be a family of functions from to .
The family is continuous if each member is continuous: For each entourage in , for each function in and each point , for some entourage in , for each point in , whenever , we have .
In short:
The family is uniformly continuous if each member is uniformly continuous: For each entourage in , for each function in , for some entourage in , for each point in , for each point in , whenever , we have .
In short:
The family is equicontinuous if: For each entourage in , for each point in , for some entourage in , for each function in , for each point in , whenever , we have .
In short:
The family is uniformly equicontinuous if: For each entourage in , for some entourage in , for each function in and each point in , for each point in , whenever , we have .
In short:
All of these definitions are identical except for the placement of the quantifiers and before or after the quantifier .
Just as it is reasonable to speak of a single function continuous at a single point , so it is reasonable to speak of a family equicontinuous at a single point (or, for that matter, of a single function uniformly continuous). However, it makes no sense to speak of a single function equicontinuous (or uniformly equicontinuous), nor can we speak of a family of functions uniformly equicontinuous at a single point .
The Cauchy sum theorem holds for an equicontinuous family of functions (from the real line to itself), without the requirement for uniform convergence.
The importance of equicontinuity is perhaps best illustrated by the Arzelà-Ascoli theorem, which gives conditions for a set in a function space to be compact. A reasonably general version is this:
Let be a convergence space and a uniform space. Then a subset is relatively compact (has compact closure) iff it is equicontinuous and is relatively compact in for each .
See BB, corollary 2.4.9. Here the topology on the space of continuous functions is the so-called natural topology, namely the largest topology on such that for all spaces , the continuity of a map implies the continuity of its transpose . This is the same as the exponential in whenever the exponential exists. See Escardó, sections 8.1 and 10.2; see also ELS where it is shown that the natural topology on coincides with the topology of continuous convergence (which is the context for the theorem above).
(Some applications of Arzela-Ascoli should also be given.)
It happens that the property of a family of functions being (uniformly) equicontinuous is equivalent to them defining a single function which is (uniformly) continuous.
Recall that for a family of functions for , we can define a function by , and that the are all (uniformly) continuous iff is (uniformly) continuous when is given the product uniformity.
However, we can give another uniformity more analogous to the box topology of topological spaces. Specifically, the box power is the uniform space with point set , and where is an entourage iff there is some entourage on such that for all in and all , . This means that the entoruages of are generated by for entourages on .
With this definition, we can see that the family of functions is (uniformly) equicontinuous iff is (uniformly) continuous.
This also gives a convenient way to show that an equicontinuous net of continuous functions which converge pointwise converges to a continuous function: a net of equicontinuous functions gives a continuous function . By the below proposition, if the converge pointwise, they converge to a continuous function.
Let be a separated uniform space and a directed set, let be the subspace of which contains the convergent nets. Then the map which takes a net to the point it converges to is continuous.
Let in and be open. Then there is an entourage on such that . is a neighbourhood of . For any , we have that (as all are in this closed set), so . This means that is continuous at , as required.
Wikipedia, Equicontinuity
R. Beattie and H.-P. Butzmann, Convergence Structures and Applications to Functional Analysis, Kluwer Academic Publishers (2002).
Last revised on November 17, 2021 at 20:33:14. See the history of this page for a list of all contributions to it.