# Difference between revisions of "Adjoint representation"

en>Pmanderson m (decap) |
en>TakuyaMurata m (→Definition) |
||

(One intermediate revision by one other user not shown) | |||

Line 1: | Line 1: | ||

{{Lie groups |Algebras}} | |||

{{mergefrom|Adjoint bundle|date=December 2014}} | |||

{{mergefrom|Adjoint representation of a Lie algebra|date=December 2014}} | |||

In [[mathematics]], the '''adjoint representation''' (or '''adjoint action''') of a [[Lie group]] ''G'' is a way of representing the elements of the group as [[linear map|linear transformations]] of the group's [[Lie algebra]], considered as a [[vector space]]. For example, in the case where ''G'' is the Lie group of [[invertible matrix|invertible matrices of size ''n'']], ''GL(n)'', the Lie algebra is the [[vector space]] of all (not necessarily invertible) ''n''-by-''n'' matrices. So in this case the adjoint representation is the vector space of ''n''-by-''n'' matrices <math> x </math>, and any element ''g'' in ''GL(n)'' [[group action|acts]] as a linear transformation of this vector space given by conjugation: <math> x \mapsto g x g^{-1} </math>. | |||

For any Lie group, this natural [[group representation|representation]] is obtained by linearizing (i.e. taking the [[Differential of a function|differential]] of) the [[group action|action]] of ''G'' on itself by [[conjugation (group theory)|conjugation]]. The adjoint representation can be defined for [[linear algebraic group]]s over arbitrary [[field (mathematics)|fields]]. | |||

{{ | ==Definition== | ||

{{see also|Representation theory|Lie_group#The Lie algebra associated with a Lie group}} | |||

Let ''G'' be a [[Lie group]] and let <math>\mathfrak g</math> be its [[Lie algebra]] (which we identify with ''T<sub>e</sub>G'', the [[tangent space]] to the [[identity element]] in ''G''). Define the map | |||

:<math>\Psi : G \to \mathrm{Aut}(G), \, g \mapsto \Psi_g</math> | |||

where Aut(''G'') is the [[automorphism group]] of ''G'' and the [[automorphism]] Ψ<sub>''g''</sub> is defined by | |||

:<math>\Psi_g(h) = ghg^{-1}\,</math> | |||

for all ''h'' in ''G''. The [[pushforward (differential)|differential]] of Ψ<sub>''g''</sub> at the identity is an automorphism of the Lie algebra <math>\mathfrak g</math>. We denote this map by Ad<sub>''g''</sub>: | |||

:<math>d(\Psi_g)_e=\mathrm{Ad}_g\colon \mathfrak g \to \mathfrak g.</math> | |||

To say that Ad<sub>''g''</sub> is a Lie algebra automorphism is to say that Ad<sub>''g''</sub> is a [[linear transformation]] of <math>\mathfrak g</math> that preserves the [[Lie algebra#Definitions|Lie bracket]]. The map | |||

:<math>\mathrm{Ad}\colon G \to \mathrm{Aut}(\mathfrak g), \, g \mapsto \mathrm{Ad}_g</math> | |||

is called the '''adjoint representation''' of ''G''. This is indeed a [[group representation|representation]] of ''G'' since <math>\mathrm{Aut}(\mathfrak g)</math> is a closed<ref>The condition that a linear map is a Lie algebra homomorphism is a closed condition.</ref> [[Lie subgroup]] of <math>\mathrm{GL}(\mathfrak g)</math> and the above adjoint map is a [[Lie group homomorphism]]. Note Ad is a trivial map if ''G'' is abelian. | |||

If ''G'' is an (immersed) Lie subgroup of the general linear group <math>GL_n(\mathbb{C})</math>, then, since the exponential map is the matrix exponential: <math>\operatorname{exp}(X) = e^X</math>, taking the derivative of <math>\Psi_g(\operatorname{exp}(tX)) = e^{t gXg^{-1}}</math> at ''t'' = 0, one gets: for ''g'' in ''G'' and ''X'' in <math>\mathfrak{g}</math>, | |||

:<math>\operatorname{Ad}_g(X) = gX g^{-1}</math> | |||

where on the right we have the products of matrices. | |||

===Adjoint representation of a Lie algebra=== | |||

{{main|Adjoint representation of a Lie algebra}} | |||

One may always pass from a representation of a Lie group ''G'' to a [[representation of a Lie algebra|representation of its Lie algebra]] by taking the derivative at the identity. | |||

Taking the derivative of the adjoint map | |||

:<math>\mathrm{Ad}\colon G \to \mathrm{Aut}(\mathfrak g)</math> | |||

gives the '''adjoint representation''' of the Lie algebra <math>\mathfrak g</math>: | |||

:<math>d(\mathrm{Ad})_x: T_x(G) \to T_{Ad(x)}(\mathrm{Aut}(\mathfrak g))</math> | |||

:<math>\mathrm{ad}\colon \mathfrak g \to \mathrm{Der}(\mathfrak g).</math> | |||

Here <math>\mathrm{Der}(\mathfrak g)</math> is the Lie algebra of <math>\mathrm{Aut}(\mathfrak g)</math> which may be identified with the [[differential algebra#Derivation on a Lie algebra|derivation algebra]] of <math>\mathfrak g</math>. The adjoint representation of a Lie algebra is related in a fundamental way to the structure of that algebra. In particular, one can show that | |||

:<math>\mathrm{ad}_x(y) = [x,y]\,</math> | |||

for all <math>x,y \in \mathfrak g</math>.<ref> | |||

<math> | |||

\begin{align} | |||

\mathrm{ad}_x(y) & = d (\mathrm{Ad}_{x})_{e}(y) \\ | |||

& = \lim_{\varepsilon \to 0}\frac{(I+\varepsilon x)y(I+\varepsilon x)^{-1}-y}{\varepsilon} \\ | |||

& = \lim_{\varepsilon \to 0}\frac{(I+\varepsilon x)y(I-\varepsilon x +(\varepsilon x)^2+O(\varepsilon^3))-y}{\varepsilon} \\ | |||

& = \lim_{\varepsilon \to 0}\frac{((I+\varepsilon x)yI- (I+\varepsilon x)y\varepsilon x +(I+\varepsilon x)y(\varepsilon x)^2 +O(\varepsilon^3))-y}{\varepsilon} \\ | |||

& = \lim_{\varepsilon \to 0}\frac{(I y I+\varepsilon x y I- I y \varepsilon x-\varepsilon x y \varepsilon x +Iy(\varepsilon x)^2+\varepsilon xy(\varepsilon x)^2 +O(\varepsilon^3))-y}{\varepsilon} \\ | |||

& = \lim_{\varepsilon \to 0}\frac{y+ x y \varepsilon - y x \varepsilon- x y x \varepsilon^{2} +y x^{2}\varepsilon^2 + x y x^{2}\varepsilon^2 +O(\varepsilon^3) -y}{\varepsilon} \\ | |||

& = \lim_{\varepsilon \to 0}x y - y x - x y x \varepsilon +y x^{2}\varepsilon + x y x^{2}\varepsilon +O(\varepsilon^2) \\ | |||

& = [x,y] | |||

\end{align} | |||

</math></ref> | |||

== Examples == | |||

*If ''G'' is [[abelian group|abelian]] of dimension ''n'', the adjoint representation of ''G'' is the trivial ''n''-dimensional representation. | |||

*If ''G'' is a [[matrix Lie group]] (i.e. a closed subgroup of ''GL''(n,'''C''')), then its Lie algebra is an algebra of ''n''×''n'' matrices with the commutator for a Lie bracket (i.e. a subalgebra of <math>\mathfrak{gl}_n(\mathbf C)</math>). In this case, the adjoint map is given by Ad<sub>''g''</sub>(''x'') = ''gxg''<sup>−1</sup>. | |||

*If ''G'' is [[SL2(R)|SL(2, '''R''')]] (real 2×2 matrices with [[determinant]] 1), the Lie algebra of ''G'' consists of real 2×2 matrices with [[trace (linear algebra)|trace]] 0. The representation is equivalent to that given by the action of ''G'' by linear substitution on the space of binary (i.e., 2 variable) [[quadratic form]]s. | |||

==Properties== | |||

The following table summarizes the properties of the various maps mentioned in the definition | |||

{| align=center border=1 cellpadding=2 style="border: solid 1pt black; border-collapse: collapse;" | |||

| align=center | <math>\Psi\colon G \to \mathrm{Aut}(G)\,</math> | |||

| align=center | <math>\Psi_g\colon G \to G\,</math> | |||

|- | |||

| valign=top | Lie group homomorphism: | |||

* <math>\Psi_{gh} = \Psi_g\Psi_h</math> | |||

| valign=top | Lie group automorphism: | |||

* <math>\Psi_g(ab) = \Psi_g(a)\Psi_g(b)</math> | |||

* <math>(\Psi_g)^{-1} = \Psi_{g^{-1}}</math> | |||

|- | |||

| align=center | <math>\mathrm{Ad}\colon G \to \mathrm{Aut}(\mathfrak g)</math> | |||

| align=center | <math>\mathrm{Ad}_g\colon \mathfrak g \to \mathfrak g</math> | |||

|- | |||

| valign=top | Lie group homomorphism: | |||

* <math>\mathrm{Ad}_{gh} = \mathrm{Ad}_g\mathrm{Ad}_h</math> | |||

| valign=top | Lie algebra automorphism: | |||

* <math>\mathrm{Ad}_g</math> is linear | |||

* <math>(\mathrm{Ad}_g)^{-1} = \mathrm{Ad}_{g^{-1}}</math> | |||

* <math>\mathrm{Ad}_g[x,y] = [\mathrm{Ad}_g x,\mathrm{Ad}_g y]</math> | |||

|- | |||

| align=center | <math>\mathrm{ad}\colon \mathfrak g \to \mathrm{Der}(\mathfrak g)</math> | |||

| align=center | <math>\mathrm{ad}_x\colon \mathfrak g \to \mathfrak g</math> | |||

|- | |||

| valign=top | Lie algebra homomorphism: | |||

* <math>\mathrm{ad}</math> is linear | |||

* <math>\mathrm{ad}_{[x,y]} = [\mathrm{ad}_x,\mathrm{ad}_y]</math> | |||

| valign=top | Lie algebra derivation: | |||

* <math>\mathrm{ad}_x</math> is linear | |||

* <math>\mathrm{ad}_x[y,z] = [\mathrm{ad}_x y ,z] + [y,\mathrm{ad}_x z]</math> | |||

|} | |||

The [[image (mathematics)|image]] of ''G'' under the adjoint representation is denoted by Ad(''G''). If ''G'' is [[connected space|connected]], the [[kernel (group theory)|kernel]] of the adjoint representation coincides with the kernel of Ψ which is just the [[center (group theory)|center]] of ''G''. Therefore the adjoint representation of a connected Lie group ''G'' is [[faithful representation|faithful]] if and only if ''G'' is centerless. More generally, if ''G'' is not connected, then the kernel of the adjoint map is the [[centralizer]] of the [[identity component]] ''G''<sub>0</sub> of ''G''. By the [[first isomorphism theorem]] we have | |||

:<math>\mathrm{Ad}(G) \cong G/Z_G(G_0).</math> | |||

Given a real Lie algebra <math>\mathfrak{g}</math>, by [[Lie's third theorem]], there is a connected Lie group <math>\operatorname{Int}(\mathfrak{g})</math> whose Lie algebra is the image of the adjoint representation of <math>\mathfrak{g}</math> (i.e., <math>\operatorname{Lie}(\operatorname{Int}(\mathfrak{g})) = \operatorname{ad}(\mathfrak{g})</math>.) It is called the '''adjoint group''' of <math>\mathfrak{g}</math>. | |||

Now, if <math>\mathfrak{g}</math> is the Lie algebra of a connected Lie group ''G'', then <math>\operatorname{Int}(\mathfrak{g})</math> is the image of the adjoint representation of ''G'': <math>\operatorname{Int}(\mathfrak{g}) = \operatorname{Ad}(G)</math>. | |||

== Roots of a semisimple Lie group == | |||

If ''G'' is [[semisimple group|semisimple]], the non-zero [[weight (representation theory)|weights]] of the adjoint representation form a [[root system]]. To see how this works, consider the case ''G'' = SL(''n'', '''R'''). We can take the group of diagonal matrices diag(''t''<sub>1</sub>, ..., ''t''<sub>''n''</sub>) as our [[maximal torus]] ''T''. Conjugation by an element of ''T'' sends | |||

:<math>\begin{bmatrix} | |||

a_{11}&a_{12}&\cdots&a_{1n}\\ | |||

a_{21}&a_{22}&\cdots&a_{2n}\\ | |||

\vdots&\vdots&\ddots&\vdots\\ | |||

a_{n1}&a_{n2}&\cdots&a_{nn}\\ | |||

\end{bmatrix} | |||

\mapsto | |||

\begin{bmatrix} | |||

a_{11}&t_1t_2^{-1}a_{12}&\cdots&t_1t_n^{-1}a_{1n}\\ | |||

t_2t_1^{-1}a_{21}&a_{22}&\cdots&t_2t_n^{-1}a_{2n}\\ | |||

\vdots&\vdots&\ddots&\vdots\\ | |||

t_nt_1^{-1}a_{n1}&t_nt_2^{-1}a_{n2}&\cdots&a_{nn}\\ | |||

\end{bmatrix}. | |||

</math> | |||

Thus, ''T'' acts trivially on the diagonal part of the Lie algebra of ''G'' and with eigenvectors ''t''<sub>''i''</sub>''t''<sub>''j''</sub><sup>−1</sup> on the various off-diagonal entries. The roots of ''G'' are the weights diag(''t''<sub>1</sub>, ..., ''t''<sub>''n''</sub>) → ''t''<sub>''i''</sub>''t''<sub>''j''</sub><sup>−1</sup>. This accounts for the standard description of the root system of ''G'' = SL<sub>''n''</sub>('''R''') as the set of vectors of the form ''e<sub>i</sub>''−''e<sub>j</sub>''. | |||

=== Example SL(2, R) === | |||

Let us compute the root system for one of the simplest cases of Lie Groups. Let us consider the group SL(2, '''R''') of two dimensional matrices with determinant 1. This consists of the set of matrices of the form: | |||

: <math>\begin{bmatrix} | |||

a & b\\ | |||

c & d\\ | |||

\end{bmatrix} </math> | |||

with ''a'', ''b'', ''c'', ''d'' real and ''ad'' − ''bc'' = 1. | |||

A maximal compact connected abelian Lie subgroup, or maximal torus ''T'', is given by the subset of all matrices of the form | |||

: <math>\begin{bmatrix} | |||

t_1 & 0\\ | |||

0 & t_2\\ | |||

\end{bmatrix} | |||

= | |||

\begin{bmatrix} | |||

t_1 & 0\\ | |||

0 & 1/t_1\\ | |||

\end{bmatrix} | |||

= | |||

\begin{bmatrix} | |||

\exp(\theta) & 0 \\ | |||

0 & \exp(-\theta) \\ | |||

\end{bmatrix} </math> | |||

with <math> t_1 t_2 = 1 </math>. The Lie algebra of the maximal torus is the Cartan subalgebra consisting of the matrices | |||

: <math> | |||

\begin{bmatrix} | |||

\theta & 0\\ | |||

0 & -\theta \\ | |||

\end{bmatrix} = | |||

\theta\begin{bmatrix} | |||

1 & 0\\ | |||

0 & 0 \\ | |||

\end{bmatrix}-\theta\begin{bmatrix} | |||

0 & 0\\ | |||

0 & 1 \\ | |||

\end{bmatrix} | |||

= \theta(e_1-e_2). | |||

</math> | |||

If we conjugate an element of SL(2, ''R'') by an element of the maximal torus we obtain | |||

: <math> | |||

\begin{bmatrix} | |||

t_1 & 0\\ | |||

0 & 1/t_1\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

a & b\\ | |||

c & d\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

1/t_1 & 0\\ | |||

0 & t_1\\ | |||

\end{bmatrix} | |||

= | |||

\begin{bmatrix} | |||

a t_1 & b t_1 \\ | |||

c / t_1 & d / t_1\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

1 / t_1 & 0\\ | |||

0 & t_1\\ | |||

\end{bmatrix} | |||

= | |||

\begin{bmatrix} | |||

a & b t_1^2\\ | |||

c t_1^{-2} & d\\ | |||

\end{bmatrix} | |||

</math> | |||

The matrices | |||

: <math> | |||

\begin{bmatrix} | |||

1 & 0\\ | |||

0 & 0\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

0 & 0\\ | |||

0 & 1\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

0 & 1\\ | |||

0 & 0\\ | |||

\end{bmatrix} | |||

\begin{bmatrix} | |||

0 & 0\\ | |||

1 & 0\\ | |||

\end{bmatrix} | |||

</math> | |||

are then 'eigenvectors' of the conjugation operation with eigenvalues <math>1,1,t_1^2, t_1^{-2}</math>. The function Λ which gives <math>t_1^2</math> is a multiplicative character, or homomorphism from the group's torus to the underlying field R. The function λ giving θ is a weight of the Lie Algebra with weight space given by the span of the matrices. | |||

It is satisfying to show the multiplicativity of the character and the linearity of the weight. It can further be proved that the differential of Λ can be used to create a weight. It is also educational to consider the case of SL(3, '''R'''). | |||

== Variants and analogues == | |||

The adjoint representation can also be defined for [[algebraic group]]s over any field. | |||

The '''co-adjoint representation''' is the [[contragredient representation]] of the adjoint representation. [[Alexandre Kirillov]] observed that the [[orbit (group theory)|orbit]] of any vector in a co-adjoint representation is a [[symplectic manifold]]. According to the philosophy in [[representation theory]] known as the '''orbit method''' (see also the [[Kirillov character formula]]), the irreducible representations of a Lie group ''G'' should be indexed in some way by its co-adjoint orbits. This relationship is closest in the case of [[nilpotent Lie group]]s. | |||

==Notes== | |||

{{reflist}} | |||

==References== | |||

*{{Fulton-Harris}} | |||

{{DEFAULTSORT:Adjoint Representation Of A Lie Group}} | |||

[[Category:Representation theory of Lie groups]] |

## Latest revision as of 00:01, 18 December 2014

Template:Lie groups
Template:Mergefrom
Template:Mergefrom
In mathematics, the **adjoint representation** (or **adjoint action**) of a Lie group *G* is a way of representing the elements of the group as linear transformations of the group's Lie algebra, considered as a vector space. For example, in the case where *G* is the Lie group of invertible matrices of size *n*, *GL(n)*, the Lie algebra is the vector space of all (not necessarily invertible) *n*-by-*n* matrices. So in this case the adjoint representation is the vector space of *n*-by-*n* matrices , and any element *g* in *GL(n)* acts as a linear transformation of this vector space given by conjugation: .

For any Lie group, this natural representation is obtained by linearizing (i.e. taking the differential of) the action of *G* on itself by conjugation. The adjoint representation can be defined for linear algebraic groups over arbitrary fields.

## Definition

{{#invoke:see also|seealso}}
Let *G* be a Lie group and let be its Lie algebra (which we identify with *T _{e}G*, the tangent space to the identity element in

*G*). Define the map

where Aut(*G*) is the automorphism group of *G* and the automorphism Ψ_{g} is defined by

for all *h* in *G*. The differential of Ψ_{g} at the identity is an automorphism of the Lie algebra . We denote this map by Ad_{g}:

To say that Ad_{g} is a Lie algebra automorphism is to say that Ad_{g} is a linear transformation of that preserves the Lie bracket. The map

is called the **adjoint representation** of *G*. This is indeed a representation of *G* since is a closed^{[1]} Lie subgroup of and the above adjoint map is a Lie group homomorphism. Note Ad is a trivial map if *G* is abelian.

If *G* is an (immersed) Lie subgroup of the general linear group , then, since the exponential map is the matrix exponential: , taking the derivative of at *t* = 0, one gets: for *g* in *G* and *X* in ,

where on the right we have the products of matrices.

### Adjoint representation of a Lie algebra

{{#invoke:main|main}}

One may always pass from a representation of a Lie group *G* to a representation of its Lie algebra by taking the derivative at the identity.

Taking the derivative of the adjoint map

gives the **adjoint representation** of the Lie algebra :

Here is the Lie algebra of which may be identified with the derivation algebra of . The adjoint representation of a Lie algebra is related in a fundamental way to the structure of that algebra. In particular, one can show that

## Examples

- If
*G*is abelian of dimension*n*, the adjoint representation of*G*is the trivial*n*-dimensional representation. - If
*G*is a matrix Lie group (i.e. a closed subgroup of*GL*(n,**C**)), then its Lie algebra is an algebra of*n*×*n*matrices with the commutator for a Lie bracket (i.e. a subalgebra of ). In this case, the adjoint map is given by Ad_{g}(*x*) =*gxg*^{−1}. - If
*G*is SL(2,**R**) (real 2×2 matrices with determinant 1), the Lie algebra of*G*consists of real 2×2 matrices with trace 0. The representation is equivalent to that given by the action of*G*by linear substitution on the space of binary (i.e., 2 variable) quadratic forms.

## Properties

The following table summarizes the properties of the various maps mentioned in the definition

Lie group homomorphism: | Lie group automorphism: |

Lie group homomorphism: | Lie algebra automorphism: |

Lie algebra homomorphism: | Lie algebra derivation: |

The image of *G* under the adjoint representation is denoted by Ad(*G*). If *G* is connected, the kernel of the adjoint representation coincides with the kernel of Ψ which is just the center of *G*. Therefore the adjoint representation of a connected Lie group *G* is faithful if and only if *G* is centerless. More generally, if *G* is not connected, then the kernel of the adjoint map is the centralizer of the identity component *G*_{0} of *G*. By the first isomorphism theorem we have

Given a real Lie algebra , by Lie's third theorem, there is a connected Lie group whose Lie algebra is the image of the adjoint representation of (i.e., .) It is called the **adjoint group** of .

Now, if is the Lie algebra of a connected Lie group *G*, then is the image of the adjoint representation of *G*: .

## Roots of a semisimple Lie group

If *G* is semisimple, the non-zero weights of the adjoint representation form a root system. To see how this works, consider the case *G* = SL(*n*, **R**). We can take the group of diagonal matrices diag(*t*_{1}, ..., *t*_{n}) as our maximal torus *T*. Conjugation by an element of *T* sends

Thus, *T* acts trivially on the diagonal part of the Lie algebra of *G* and with eigenvectors *t*_{i}*t*_{j}^{−1} on the various off-diagonal entries. The roots of *G* are the weights diag(*t*_{1}, ..., *t*_{n}) → *t*_{i}*t*_{j}^{−1}. This accounts for the standard description of the root system of *G* = SL_{n}(**R**) as the set of vectors of the form *e _{i}*−

*e*.

_{j}### Example SL(2, R)

Let us compute the root system for one of the simplest cases of Lie Groups. Let us consider the group SL(2, **R**) of two dimensional matrices with determinant 1. This consists of the set of matrices of the form:

with *a*, *b*, *c*, *d* real and *ad* − *bc* = 1.

A maximal compact connected abelian Lie subgroup, or maximal torus *T*, is given by the subset of all matrices of the form

with . The Lie algebra of the maximal torus is the Cartan subalgebra consisting of the matrices

If we conjugate an element of SL(2, *R*) by an element of the maximal torus we obtain

The matrices

are then 'eigenvectors' of the conjugation operation with eigenvalues . The function Λ which gives is a multiplicative character, or homomorphism from the group's torus to the underlying field R. The function λ giving θ is a weight of the Lie Algebra with weight space given by the span of the matrices.

It is satisfying to show the multiplicativity of the character and the linearity of the weight. It can further be proved that the differential of Λ can be used to create a weight. It is also educational to consider the case of SL(3, **R**).

## Variants and analogues

The adjoint representation can also be defined for algebraic groups over any field.

The **co-adjoint representation** is the contragredient representation of the adjoint representation. Alexandre Kirillov observed that the orbit of any vector in a co-adjoint representation is a symplectic manifold. According to the philosophy in representation theory known as the **orbit method** (see also the Kirillov character formula), the irreducible representations of a Lie group *G* should be indexed in some way by its co-adjoint orbits. This relationship is closest in the case of nilpotent Lie groups.