# Topological group

In mathematics, **topological groups** are logically the combination of groups and topological spaces, i.e. they are group and topological spaces at the same time, s.t. the continuity condition for the group operations connect these two structures together and consequently they are not independent from each other.^{[1]}

Topological groups have been studied extensively in the period of 1925 to 1940. Haar and Weil (respectively in 1933 and 1940) showed that the integrals and Fourier series are special cases of a very wide class of topological groups.^{[2]}

Topological groups, along with continuous group actions, are used to study continuous symmetries, which have many applications, for example, in physics. In functional analysis, every topological vector space is an additive topological group with the additional property that scalar multiplication is continuous; consequently, many results from the theory of topological groups can be applied to functional analysis.

A **topological group**, G, is a topological space that is also a group such that the group operation (in this case product):

are continuous^{[note 1]} Here *G* × *G* is viewed as a topological space with the product topology. Such a topology is said to be **compatible with the group operations** and is called a **group topology**.

To show that a topology is compatible with the group operations, it suffices to check that the map

is continuous. Explicitly, this means that for any *x*, *y* ∈ *G* and any neighborhood W in G of *xy* ^{−1}, there exist neighborhoods U of x and V of y in G such that *U* ⋅ (*V* ^{−1}) ⊆ *W*.

This definition used notation for multiplicative groups; the equivalent for additive groups would be that the following two operations are continuous:

Although not part of this definition, many authors^{[3]} require that the topology on G be Hausdorff. One reason for this is that any topological group can be canonically associated with a Hausdorff topological group by taking an appropriate canonical quotient; this however, often still requires working with the original non-Hausdorff topological group.
Other reasons, and some equivalent conditions, are discussed below.

This article will not assume that topological groups are necessarily Hausdorff.

In the language of category theory, topological groups can be defined concisely as group objects in the category of topological spaces, in the same way that ordinary groups are group objects in the category of sets. Note that the axioms are given in terms of the maps (binary product, unary inverse, and nullary identity), hence are categorical definitions.

A **homomorphism** of topological groups means a continuous group homomorphism *G* → *H*. Topological groups, together with their homomorphisms, form a category. A group homomorphism between commutative topological groups is continuous if and only if it is continuous at *some* point.^{[4]}

An **isomorphism** of topological groups is a group isomorphism that is also a homeomorphism of the underlying topological spaces. This is stronger than simply requiring a continuous group isomorphism—the inverse must also be continuous. There are examples of topological groups that are isomorphic as ordinary groups but not as topological groups. Indeed, any non-discrete topological group is also a topological group when considered with the discrete topology. The underlying groups are the same, but as topological groups there is not an isomorphism.

Every group can be trivially made into a topological group by considering it with the discrete topology; such groups are called discrete groups. In this sense, the theory of topological groups subsumes that of ordinary groups. The indiscrete topology (i.e. the trivial topology) also makes every group into a topological group.

The real numbers, ℝ with the usual topology form a topological group under addition. Euclidean n-space ℝ^{n} is also a topological group under addition, and more generally, every topological vector space forms an (abelian) topological group. Some other examples of abelian topological groups are the circle group *S*^{1}, or the torus (*S*^{1})^{n} for any natural number n.

The classical groups are important examples of non-abelian topological groups. For instance, the general linear group GL(*n*,ℝ) of all invertible n-by-n matrices with real entries can be viewed as a topological group with the topology defined by viewing GL(*n*,ℝ) as a subspace of Euclidean space ℝ^{n×n}. Another classical group is the orthogonal group O(*n*), the group of all linear maps from ℝ^{n} to itself that preserve the length of all vectors. The orthogonal group is compact as a topological space. Much of Euclidean geometry can be viewed as studying the structure of the orthogonal group, or the closely related group *O*(*n*) ⋉ ℝ^{n} of isometries of ℝ^{n}.

The groups mentioned so far are all Lie groups, meaning that they are smooth manifolds in such a way that the group operations are smooth, not just continuous. Lie groups are the best-understood topological groups; many questions about Lie groups can be converted to purely algebraic questions about Lie algebras and then solved.

An example of a topological group that is not a Lie group is the additive group ℚ of rational numbers, with the topology inherited from ℝ. This is a countable space, and it does not have the discrete topology. An important example for number theory is the group ℤ_{p} of *p*-adic integers, for a prime number p, meaning the inverse limit of the finite groups ℤ/*p*^{n} as *n* goes to infinity. The group ℤ_{p} is well behaved in that it is compact (in fact, homeomorphic to the Cantor set), but it differs from (real) Lie groups in that it is totally disconnected. More generally, there is a theory of *p*-adic Lie groups, including compact groups such as GL(*n*,ℤ_{p}) as well as locally compact groups such as GL(*n*,ℚ_{p}), where ℚ_{p} is the locally compact field of *p*-adic numbers.

Some topological groups can be viewed as infinite dimensional Lie groups; this phrase is best understood informally, to include several different families of examples. For example, a topological vector space, such as a Banach space or Hilbert space, is an abelian topological group under addition. Some other infinite-dimensional groups that have been studied, with varying degrees of success, are loop groups, Kac–Moody groups, diffeomorphism groups, homeomorphism groups, and gauge groups.

In every Banach algebra with multiplicative identity, the set of invertible elements forms a topological group under multiplication. For example, the group of invertible bounded operators on a Hilbert space arises this way.

The inversion operation on a topological group G is a homeomorphism from G to itself. Likewise, if *a* is any element of G, then left or right multiplication by *a* yields a homeomorphism *G* → *G*. Consequently, if 𝒩 is a neighborhood basis of the identity element in a commutative topological group G then for all *x* ∈ *X*,

A subset S of G is said to be symmetric if *S* ^{−1} = *S*. The closure of every symmetric set in a commutative topological group is symmetric.^{[4]} If S is any subset of a commutative topological group G, then the following sets are also symmetric: *S* ^{−1} ∩ *S*, *S* ^{−1} ∪ *S*, and *S* ^{−1} *S*.^{[4]}

For any neighborhood N in a commutative topological group G of the identity element, there exists a symmetric neighborhood M of the identity element such that *M* ^{−1} *M* ⊆ *N*, where note that *M* ^{−1} *M* is necessarily a symmetric neighborhood of the identity element.^{[4]} Thus every topological group has a neighborhood basis at the identity element consisting of symmetric sets.

If G is a locally compact commutative group, then for any neighborhood N in G of the identity element, there exists a symmetric relatively compact neighborhood M of the identity element such that cl *M* ⊆ *N* (where cl *M* is symmetric as well).^{[4]}

Every topological group can be viewed as a uniform space in two ways; the *left uniformity* turns all left multiplications into uniformly continuous maps while the *right uniformity* turns all right multiplications into uniformly continuous maps.^{[5]}
If G is not abelian, then these two need not coincide. The uniform structures allow one to talk about notions such as completeness, uniform continuity and uniform convergence on topological groups.

If U is an open subset of a commutative topological group G and U contains a compact set K, then there exists a neighborhood N of the identity element such that *KN* ⊆ *U*.^{[4]}

As a uniform space, every commutative topological group is completely regular. Consequently, for a multiplicative topological group G with identity element 1, the following are equivalent:^{[4]}

A subgroup of a commutative topological group is discrete if and only if it has an isolated point.^{[4]}

If G is not Hausdorff, then one can obtain a Hausdorff group by passing to the quotient group *G*/*K*, where K is the closure of the identity.^{[6]} This is equivalent to taking the Kolmogorov quotient of G.

The **Birkhoff–Kakutani theorem** (named after mathematicians Garrett Birkhoff and Shizuo Kakutani) states that the following three conditions on a topological group G are equivalent:^{[7]}

(A metric on G is called left-invariant if for each point *a* ∈ *G*, the map *x* ↦ *ax* is an isometry from G to itself.)

Every subgroup of a topological group is itself a topological group when given the subspace topology. Every open subgroup H is also closed in G, since the complement of H is the open set given by the union of open sets *gH* for *g* ∈ *G* \ *H*. If H is a subgroup of G then the closure of H is also a subgroup. Likewise, if H is a normal subgroup of G, the closure of H is normal in G.

If H is a subgroup of G, the set of left cosets *G*/*H* with the quotient topology is called a homogeneous space for G. The quotient map *q* : *G* → *G*/*H* is always open. For example, for a positive integer n, the sphere *S*^{n} is a homogeneous space for the rotation group SO(*n*+1) in ℝ^{n+1}, with *S*^{n} = SO(*n*+1)/SO(*n*). A homogeneous space *G*/*H* is Hausdorff if and only if H is closed in G.^{[8]} Partly for this reason, it is natural to concentrate on closed subgroups when studying topological groups.

If H is a normal subgroup of G, then the quotient group *G*/*H* becomes a topological group when given the quotient topology. It is Hausdorff if and only if H is closed in G. For example, the quotient group ℝ/ℤ is isomorphic to the circle group *S*^{1}.

In any topological group, the identity component (i.e., the connected component containing the identity element) is a closed normal subgroup. If C is the identity component and *a* is any point of G, then the left coset *aC* is the component of G containing *a*. So the collection of all left cosets (or right cosets) of C in G is equal to the collection of all components of G. It follows that the quotient group *G*/*C* is totally disconnected.^{[9]}

In any commutative topological group, the product (assuming the group is multiplicative) *KC* of a compact set K and a closed set C is a closed set.^{[4]} Furthermore, for any subsets R and S of G, (cl *R*)(cl *S*) ⊆ cl (*RS*).^{[4]}

If H is a subgroup of a commutative topological group G and if N is a neighborhood in G of the identity element such that *H* ∩ cl *N* is closed, then H is closed.^{[4]} Every discrete subgroup of a Hausdorff commutative topological group is closed.^{[4]}

The isomorphism theorems from ordinary group theory are not always true in the topological setting. This is because a bijective homomorphism need not be an isomorphism of topological groups. The theorems are valid if one places certain restrictions on the maps involved. For example, the first isomorphism theorem states that if *f* : *G* → *H* is a homomorphism, then the homomorphism from *G*/ker(*f*) to im(*f*) is an isomorphism if and only if the map f is open onto its image.^{[10]}

**Hilbert's fifth problem** asked whether a topological group G that is a topological manifold must be a Lie group. In other words, does G have the structure of a smooth manifold, making the group operations smooth? As shown by Andrew Gleason, Deane Montgomery, and Leo Zippin, the answer to this problem is yes.^{[11]} In fact, G has a real analytic structure. Using the smooth structure, one can define the Lie algebra of G, an object of linear algebra that determines a connected group G up to covering spaces. As a result, the solution to Hilbert's fifth problem reduces the classification of topological groups that are topological manifolds to an algebraic problem, albeit a complicated problem in general.

The theorem also has consequences for broader classes of topological groups. First, every compact group (understood to be Hausdorff) is an inverse limit of compact Lie groups. (One important case is an inverse limit of finite groups, called a profinite group. For example, the group ℤ_{p} of *p*-adic integers and the absolute Galois group of a field are profinite groups.) Furthermore, every connected locally compact group is an inverse limit of connected Lie groups.^{[12]} At the other extreme, a totally disconnected locally compact group always contains a compact open subgroup, which is necessarily a profinite group.^{[13]} (For example, the locally compact group GL(*n*,ℚ_{p}) contains the compact open subgroup GL(*n*,ℤ_{p}), which is the inverse limit of the finite groups GL(*n*,ℤ/*p*^{r}) as r' goes to infinity.)

An **action** of a topological group G on a topological space *X* is a group action of G on *X* such that the corresponding function *G* × *X* → *X* is continuous. Likewise, a **representation** of a topological group G on a real or complex topological vector space *V* is a continuous action of G on *V* such that for each *g* ∈ *G*, the map *v* ↦ *gv* from *V* to itself is linear.

Group actions and representation theory are particularly well understood for compact groups, generalizing what happens for finite groups. For example, every finite-dimensional (real or complex) representation of a compact group is a direct sum of irreducible representations. An infinite-dimensional unitary representation of a compact group can be decomposed as a Hilbert-space direct sum of irreducible representations, which are all finite-dimensional; this is part of the Peter–Weyl theorem.^{[14]} For example, the theory of Fourier series describes the decomposition of the unitary representation of the circle group *S*^{1} on the complex Hilbert space L^{2}(*S*^{1}). The irreducible representations of *S*^{1} are all 1-dimensional, of the form *z* ↦ *z*^{n} for integers n (where *S*^{1} is viewed as a subgroup of the multiplicative group ℂ*). Each of these representations occurs with multiplicity 1 in L^{2}(*S*^{1}).

The irreducible representations of all compact connected Lie groups have been classified. In particular, the character of each irreducible representation is given by the Weyl character formula.

More generally, locally compact groups have a rich theory of harmonic analysis, because they admit a natural notion of measure and integral, given by the Haar measure. Every unitary representation of a locally compact group can be described as a direct integral of irreducible unitary representations. (The decomposition is essentially unique if G is of Type I, which includes the most important examples such as abelian groups and semisimple Lie groups.^{[15]}) A basic example is the Fourier transform, which decomposes the action of the additive group ℝ on the Hilbert space L^{2}(ℝ) as a direct integral of the irreducible unitary representations of ℝ. The irreducible unitary representations of ℝ are all 1-dimensional, of the form *x* ↦ *e*^{2πiax} for *a* ∈ ℝ.

The irreducible unitary representations of a locally compact group may be infinite-dimensional. A major goal of representation theory, related to the Langlands classification of admissible representations, is to find the unitary dual (the space of all irreducible unitary representations) for the semisimple Lie groups. The unitary dual is known in many cases such as SL(2,ℝ), but not all.

Every locally compact group G has a good supply of irreducible unitary representations; for example, enough representations to distinguish the points of G (the Gelfand–Raikov theorem). By contrast, representation theory for topological groups that are not locally compact has so far been developed only in special situations, and it may not be reasonable to expect a general theory. For example, there are many abelian Banach–Lie groups for which every representation on Hilbert space is trivial.^{[16]}

Topological groups are special among all topological spaces, even in terms of their homotopy type. One basic point is that a topological group G determines a path-connected topological space, the classifying space *BG* (which classifies principal G-bundles over topological spaces, under mild hypotheses). The group G is isomorphic in the homotopy category to the loop space of *BG*; that implies various restrictions on the homotopy type of G.^{[17]} Some of these restrictions hold in the broader context of H-spaces.

For example, the fundamental group of a topological group G is abelian. (More generally, the Whitehead product on the homotopy groups of G is zero.) Also, for any field *k*, the cohomology ring *H**(*G*,*k*) has the structure of a Hopf algebra. In view of structure theorems on Hopf algebras by Heinz Hopf and Armand Borel, this puts strong restrictions on the possible cohomology rings of topological groups. In particular, if G is a path-connected topological group whose rational cohomology ring *H**(*G*,ℚ) is finite-dimensional in each degree, then this ring must be a free graded-commutative algebra over ℚ, that is, the tensor product of a polynomial ring on generators of even degree with an exterior algebra on generators of odd degree.^{[18]}

In particular, for a connected Lie group G, the rational cohomology ring of G is an exterior algebra on generators of odd degree. Moreover, a connected Lie group G has a maximal compact subgroup *K*, which is unique up to conjugation, and the inclusion of *K* into G is a homotopy equivalence. So describing the homotopy types of Lie groups reduces to the case of compact Lie groups. For example, the maximal compact subgroup of SL(2,ℝ) is the circle group SO(2), and the homogeneous space SL(2,ℝ)/SO(2) can be identified with the hyperbolic plane. Since the hyperbolic plane is contractible, the inclusion of the circle group into SL(2,ℝ) is a homotopy equivalence.

Information about convergence of nets and filters, such as definitions and properties, can be found in the article about filters in topology.

We will henceforth assume that any topological group that we consider is an additive commutative topological group with identity element 0.

and for any *N* ⊆ *X* containing 0, the **canonical entourage** or **canonical vicinities around N** is the set

**Definition** (**Canonical uniformity**):^{[19]} For a topological group (*X*, τ), the **canonical uniformity** on X is the uniform structure induced by the set of all canonical entourages Δ(*N*) as N ranges over all neighborhoods of 0 in X.

That is, it is the upward closure of the following prefilter on *X* × *X*,

The general theory of uniform spaces has its own definition of a "Cauchy prefilter" and "Cauchy net." For the canonical uniformity on X, these reduces down to the definition described below.

and similarly their **difference** is defined to be the image of the product net under the subtraction map:

or equivalently, if for every neighborhood N of 0 in X, there exists some *i*_{0} ∈ *I* such that *x*_{i} - *x*_{j} ∈ *N* for all *i*, *j* ≥ *i*_{0} with *i*, *j* ∈ *I*.

**Definition** (**Cauchy prefilter**): A prefilter ℬ on an additive topological group X called a **Cauchy prefilter** if it satisfies any of the following equivalent conditions:

Various generalizations of topological groups can be obtained by weakening the continuity conditions:^{[25]}