Compact group

From Wikipedia, the free encyclopedia

In mathematics, a compact (topological, often understood) group is a topological group that is also a compact space. Compact groups are a natural generalisation of finite groups with the discrete topology and have properties that carry over in significant fashion. Compact groups have a well-understood theory, in relation to group actions and representation theory.

In the following we will assume all groups are Hausdorff, since this covers all interesting cases.

1 Compact Lie groups
- 1.1 Classification
2 Further examples
3 Haar measure
4 Representation theory
5 Duality
6 From compact to non-compact groups
7 See also

[edit] Compact Lie groups

Lie groups form the nicest class of topological groups, and the compact Lie groups have a particularly well-developed theory. Basic examples of compact Lie groups include

any finite group with the discrete topology
the circle group T and the torus groups Tⁿ,
the orthogonal groups O(n), the special orthogonal group SO(n) and its covering spin group Spin(n),
the unitary group U(n) and the special unitary group SU(n),
the symplectic group Sp(n)
the compact forms of the exceptional Lie groups: G₂, F₄, E₆, E₇, and E₈.

The classification theorem of compact Lie groups states that up to finite extensions and finite covers this exhausts the list of examples (which already includes some redundancies).

[edit] Classification

Given any compact Lie group G one can take its identity component G₀ which is connected. The quotient group G/G₀ is the group of components π₀(G) which must be finite since G is compact. We therefore have a finite extension

$1\to G_0 \to G \to \pi_0(G) \to 1\,$

Now every compact, connected Lie group G₀ has a finite covering

$1\to A\to \tilde{G}_0\to G_0\to 1\,$

where $A\sub Z(\tilde{G})$ is a finite abelian group and $\tilde{G}$ is a product of a torus and a compact, connected, simply-connected Lie group K:

$\tilde{G}_0 \cong \mathbb T^m \times K$

Finally, every compact, connected, simply-connected Lie group K is a product of compact, connected, simply-connected simple Lie groups K_i each of which is isomorphic to exactly one of

Sp(n), n ≥ 1
SU(n), n ≥ 3
Spin(n), n ≥ 7
G₂, F₄, E₆, E₇, or E₈

[edit] Further examples

Amongst groups that are not Lie groups, and so do not carry the structure of a manifold, examples are the additive group Z_p of p-adic integers, and constructions from it. In fact any profinite group is a compact group. This means that Galois groups are compact groups, a basic fact for the theory of algebraic extensions in the case of infinite degree.

Pontryagin duality provides a large supply of examples of compact commutative groups. These are in duality with abelian discrete groups.

[edit] Haar measure

Compact groups all carry a Haar measure, which will be invariant by both left and right translation (the modulus function must be a continuous homomorphism to the positive multiplicative reals, and so 1). In other words these groups are unimodular. Haar measure is easily normalised to be a probability measure, analogous to dθ/2π on the circle.

Such a Haar measure is in many cases easy to compute; for example for orthogonal groups it was known to Hurwitz, and in the Lie group cases can always be given by an invariant differential form. In the profinite case there are many subgroups of finite index, and Haar measure of a coset will be the reciprocal of the index. Therefore integrals are often computable quite directly, a fact applied constantly in number theory.

[edit] Representation theory

The representation theory of compact groups was founded by the Peter-Weyl theorem. Hermann Weyl went on to give the detailed character theory of the compact connected Lie groups, based on maximal torus theory. The resulting Weyl character formula was one of the influential results of twentieth century mathematics.

A combination of Weyl's work and Cartan's theorem gives a survey of the whole representation theory of compact groups G . That is, by the Peter-Weyl theorem the irreducible unitary representations ρ of G are into a unitary group (of finite dimension) and the image will be a closed subgroup of the unitary group by compactness. Cartan's theorem states that Im(ρ) must itself be a Lie subgroup in the unitary group. If G is not itself a Lie group, there must be a kernel to ρ. Further one can form an inverse system, for the kernel of ρ smaller and smaller, of finite-dimensional unitary representations, which identifies G as an inverse limit of compact Lie groups. Here the fact that in the limit a faithful representation of G is found is another consequence of the Peter-Weyl theorem,

The unknown part of the representation theory of compact groups is thereby, roughly speaking, thrown back onto the complex representations of finite groups. This theory is rather rich in detail, but is qualitatively well understood.

[edit] Duality

The topic of recovering a compact group from its representation theory is the subject of the Tannaka-Krein duality, now often recast in term of tannakian category theory.

[edit] From compact to non-compact groups

The influence of the compact group theory on non-compact groups was formulated by Weyl in his unitarian trick. Inside a general semisimple Lie group there is a maximal compact subgroup, and the representation theory of such groups, developed largely by Harish-Chandra, uses intensively the restriction of a representation to such a subgroup, and also the model of Weyl's character theory.