Next Article in Journal
Rational Approximation for Solving an Implicitly Given Colebrook Flow Friction Equation
Previous Article in Journal
Odd Cycles and Hilbert Functions of Their Toric Rings
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Twisted Weyl Groups of Compact Lie Groups and Nonabelian Cohomology

School of Mathematical Sciences, Tiangong University, Tianjin 300387, China
*
Author to whom correspondence should be addressed.
These authors contributed equally to this work.
Mathematics 2020, 8(1), 21; https://doi.org/10.3390/math8010021
Submission received: 10 November 2019 / Revised: 15 December 2019 / Accepted: 17 December 2019 / Published: 20 December 2019

Abstract

:
Given a compact connected Lie group G with an S 1 -module structure and a maximal compact torus T of G S 1 , we define twisted Weyl group W ( G , S 1 , T ) of G associated to S 1 -module and show that two elements of T are δ -conjugate if and only if they are in one W ( G , S 1 , T ) -orbit. Based on this, we prove that the natural map W ( G , S 1 , T ) \ H 1 ( S 1 , T ) H 1 ( S 1 , G ) is bijective, which reduces the calculation for the nonabelian cohomology H 1 ( S 1 , G ) .

1. Introduction

Let G be a compact connected Lie group and σ be an automorphism on G [1]. The twisted conjugate action τ σ [2] of G on itself associated to σ is defined as
τ σ : G × G G , ( g , h ) g h σ ( g ) 1 τ g σ ( h ) .
Two elements g 1 and g 2 of G are called σ -conjugate if they are in one twisted orbit of G associated to σ , i.e., there exists an element g 0 G such that
g 2 = τ g 0 σ ( g 1 ) .
Based on this, given a compact connected Lie group G with an additional Z n -module structure and T a maximal compact torus of G 0 Z n , An J. [3] defined the twisted Weyl group W ( G , Z n , T ) of G associated to Z n -module and reduced the calculation of H 1 ( Z n , G ) to the action of W ( G , Z n , T ) on T, where H 1 ( Z n , G ) is the first nonabelian cohomology of Z n with coefficients in G [2].
Motivated by the underlying work, in this paper we consider the case of G with an nonabelian S 1 -module structure, where S 1 is the one-dimensional torus. Picking a topological generator δ of S 1 , which can be regarded as an automorphism of G [4], we first define the twisted Weyl group W ( G , δ , T ) of G associated to δ and then define the twisted Weyl group W ( G , S 1 , T ) associated to S 1 -module, where T is a maximal compact torus in G S 1 . Closed by the definition, we study the action of W ( G , S 1 , T ) on T and show that two elements in T are δ -conjugate if and only if they are in one W ( G , S 1 , T ) -orbit. Furthermore, we prove that W ( G , S 1 , T ) is a finite group, which is the same as the case of classical Weyl groups.
Based on the underlying properties of W ( G , S 1 , T ) , we study the action of W ( G , S 1 , T ) on the first cohomology H 1 ( S 1 , T ) of the compact Lie group S 1 with coefficients in T [5], and prove that the natural map
W ( G , S 1 , T ) \ H 1 ( S 1 , T ) H 1 ( S 1 , G )
induced by the natural embedding T G is a bijection. Using the result, one can reduce the calculation for H 1 ( S 1 , G ) to the calculation for the orbit space W ( G , S 1 , T ) \ H 1 ( S 1 , T ) . Indeed, using this formula, one can simplify some calculations in dynamical systems theory, especially in fractional dynamics and fractional-wavelet analysis of some positive definite distributions [6,7,8].
In Section 2, we exhibit the definition of the twisted Weyl group W ( G , S 1 , T ) of G associated to S 1 -module together with its some properties. In Section 3, we construct a one-to-one correspondence between the orbit space of the action of W ( G , S 1 , T ) on H 1 ( S 1 , T ) and H 1 ( S 1 , G ) . In Section 4, we discuss some new developments in the field as well as its relations with amenability of groups [9,10,11,12]. For basic knowledge on compact Lie groups and twisted conjugate actions, one can refer [2,13,14]; for the nonabelian cohomology of Lie groups, one can refer [5,15,16,17].

2. Twisted Weyl Groups of Compact Lie Groups

Let G be a compact connected Lie group with an S 1 -module structure and δ be a topological generator of S 1 . From the definition of S 1 -module structure, δ can be regarded as an automorphism on G. Denote by τ δ the twisted conjugate action of G associated to δ .
First of all, we define the twisted Weyl group of G associated to δ .
Definition 1.
Define
N δ , G ( T ) : = { g G τ g δ ( T ) = T } ,
Z δ , G ( T ) : = { g G τ g δ ( t ) = t , t T } ,
where T is a maximal compact torus of G δ : = { g G δ ( g ) = g } .
It is easy to know that N δ , G ( T ) and Z δ , G ( T ) are both closed subgroups of G and that Z δ , G ( T ) is a normal subgroup of N δ , G ( T ) .
Definition 2.
Define
W ( G , δ , T ) : = N δ , G ( T ) / Z δ , G ( T ) .
W ( G , δ , T ) is called the twisted Weyl group of G associated to δ.
As an abstract group, the group operations on W ( G , δ , T ) are defined as following. For all w 1 , w 2 W ( G , δ , T ) ,
w 1 · w 2 : = [ g 1 g 2 ] ,
where g 1 w 1 , g 2 w 2 , [ g 1 g 2 ] represents the equivalence class of g 1 g 2 in N δ , G ( T ) / Z δ , G ( T ) ; for all w W ( G , δ , T ) ,
w 1 : = [ g 1 ] ,
where g w .
Next, we exhibit the definition of the twisted Weyl group of G associated to S 1 -module.
Definition 3.
Define,
N S 1 ( T ) : = { g G τ g δ ( T ) = T , δ S 1 } ,
Z S 1 ( T ) : = { g G τ g δ ( t ) = t , t T , δ S 1 } ,
where T is a maximal compact torus of G S 1 : = { g G a ( g ) = g , a S 1 } .
It is clear that they are both closed subgroups of G and that Z S 1 ( T ) is a normal subgroup of N S 1 ( T ) .
Proposition 1.
Let δ be a topological generator of S 1 . Then,
G S 1 = G δ .
Proof. 
It suffices to prove G S 1 G δ . Let g G δ and set
S g 1 : = { δ S 1 δ ( g ) = g } .
For the underlying g G δ , it is obvious that δ ( g ) = g . Thus, for all n Z ,
δ n ( g ) = δ n 1 ( δ ( g ) ) = δ n 1 ( g ) = = g .
Since δ is a topological generator of S 1 , S g 1 is dense in S 1 . Again S g 1 is a closed subgroup of S 1 ,
S g 1 = S 1 .
Then, a ( g ) = g , a S 1 , g G S 1 . So,
G S 1 G δ .
 □
By Proposition 1, G S 1 is connected. For the underlying two subgroups N S 1 ( T ) and Z S 1 ( T ) of G, we claim that:
Proposition 2.
For a given topological generator δ 0 of S 1 and a maximal compact torus T of G S 1 ,
(i) 
N S 1 ( T ) = N δ 0 , G ( T ) ;
(ii) 
Z S 1 ( T ) = Z δ 0 , G ( T ) .
Proof. 
(i) It suffices to show that N S 1 ( T ) N δ 0 , G ( T ) .
Suppose that g 0 N δ 0 , G ( T ) . Then
g 0 T δ 0 ( g 0 ) 1 = T .
Define
S g 0 , T 1 : = { δ S 1 g 0 T δ ( g 0 ) 1 = T } .
Then, S g 0 , T 1 is a dense subset of S 1 . In fact, for all n Z ,
g 0 T δ 0 n ( g 0 ) 1 = g 0 T δ 0 n ( g 0 1 ) = ( g 0 T δ 0 ( g 0 1 ) ) ( δ 0 ( g 0 ) T δ 0 2 ( g 0 1 ) ) ( δ 0 n 1 ( g 0 ) T δ 0 n ( g 0 1 ) ) = ( g 0 T δ 0 ( g 0 1 ) ) δ ( g 0 T δ 0 ( g 0 1 ) ) δ n 1 ( g 0 T δ 0 ( g 0 1 ) ) = T n = T .
Since δ 0 is a topological generator of S 1 , the set which is generated by δ 0 is a dense subset of S 1 . Again for all n Z ,
g 0 T δ 0 n ( g 0 ) 1 = T ,
S g 0 , T 1 is dense in S 1 . It is obvious that S g 0 , T 1 is also a closed subgroup of S 1 . Hence,
S g 0 , T 1 = S 1 ,
which shows that
τ g 0 δ ( T ) = T ,
for all δ S 1 . So, g 0 N S 1 ( T ) , i.e.,
N S 1 ( T ) N δ 0 , G ( T ) .
(ii) Similarly, it suffices to show that Z S 1 ( T ) Z δ 0 , G ( T ) .
Suppose that g 0 Z δ 0 , G ( T ) . Then we have
g 0 t δ 0 ( g 0 ) 1 = t ,
for all t T .
For g 0 Z δ 0 , G ( T ) , for any t T , define
S g 0 , t 1 : = { δ S 1 g 0 t δ ( g 0 ) 1 = t } .
Analogously,
S g 0 , t 1 = S 1 .
So,
Z S 1 ( T ) Z δ 0 , G ( T ) .
 □
From Proposition 2, one can get that the subgroups N S 1 ( T ) and Z S 1 ( T ) of G are independent with the choice of the topological generator δ of S 1 .
Definition 4.
Let δ be a topological generator of S 1 and W ( G , δ , T ) be defined as above. Define,
W ( G , S 1 , T ) : = W ( G , δ , T ) .
W ( G , S 1 , T ) is called the twisted Weyl group of G associated to S 1 -module.
Following, we present some properties of W ( G , S 1 , T ) .
Proposition 3.
W ( G , S 1 , T ) is independent with the choice of T.
Proof. 
Let T is another maximal compact torus of G δ . Then by ([3], Proposition 2.11), there exists an element g 0 G δ such that T = g 0 T g 0 1 . Thus,
τ g 0 g g 0 1 δ ( T ) = g 0 g g 0 1 T δ ( g 0 g g 0 1 ) 1 = g 0 g g 0 1 T g 0 δ ( g ) 1 g 0 1 = g 0 g T δ ( g ) 1 g 0 1 = g 0 T g 0 1 = T ,
for all g N δ , G ( T ) . Hence,
g 0 N δ , G ( T ) g 0 1 N δ , G ( T ) .
Similarly, we have
g 0 N δ , G ( T ) g 0 1 N δ , G ( T ) .
Therefore,
N δ , G ( T ) = g 0 N δ , G ( T ) g 0 1 .
Analogously,
Z δ , G ( T ) = g 0 Z δ , G ( T ) g 0 1 .
Then,
W δ ( T ) = N δ , G ( T ) / Z δ , G ( T ) = g 0 N δ , G ( T ) g 0 1 / g 0 Z δ , G ( T ) g 0 1 W δ ( T ) ,
which shows that W ( G , S 1 , T ) is independent with the choice of T. □
For the reason, we write W ( G , S 1 , T ) as W ( S 1 ) and write W ( G , δ , T ) as W ( δ ) in simplified forms, respectively.
Lemma 1.
Denote by L Lie functor. Then,
L ( Z δ , G ( T ) ) = L ( N δ , G ( T ) ) = L ( T ) .
Proof. 
Above all, we show the first equality. It suffices to show the reverse inclusion for the clear fact that L ( Z δ , G ( T ) ) L ( N δ , G ( T ) ) . For all X L ( N δ , G ( T ) ) , s R , t T , we have
τ e s X ( t ) = e s X t δ ( e s X ) 1 = e s X t e s d δ ( X ) = e s X e s A d ( t ) d δ ( X ) t T .
Then, ( 1 A d ( t ) d δ ) ( X ) L ( T ) . Now, we show
( 1 A d ( t ) d δ ) ( X ) = 0 ,
for all t T . In fact, L ( T ) k e r ( 1 A d ( t ) d δ ) implies that ( 1 A d ( t ) d δ ) 2 ( X ) = 0 . Since d δ and A d ( t ) are both 1 semisimple (see in [3]), A d ( t ) d δ is 1 semisimple. Hence, k e r ( 1 A d ( t ) d δ ) = k e r ( 1 A d ( t ) d δ ) 2 . Then, we have ( 1 A d ( t ) d δ ) ( X ) = 0 for all t T .
Next, we show the equality L ( Z δ , G ( T ) ) = L ( T ) . For all g Z δ , G ( T ) , t T , τ g δ ( t ) = t . Take e for the unit of T. Then τ g ( e ) = e , i.e., g δ ( g ) 1 = e . Thus, δ ( g ) = g and hence g G δ . So, Z δ , G ( T ) G δ . Moreover,
Z δ , G ( T ) = Z δ , G ( T ) G δ = Z δ , G δ ( T ) = T ,
which shows Z δ , G ( T ) = T . Then,
L ( Z δ , G ( T ) ) = L ( T ) .
Therefore, L ( Z δ , G ( T ) ) = L ( N δ , G ( T ) ) = L ( T ) .  □
Proposition 4.
As an abstract group, W ( G , S 1 , T ) is a finite group.
Proof. 
By Lemma 1, it is clear. □
Remark 1.
In the case of classical Weyl groups, let G be a compact connected Lie group, T be a maximal torus of G, and g 0 and t 0 be the Lie algebras of G and T respectively. Denote by g and t the complexifications of g 0 and t 0 , and denote Δ ( g , t ) be the set of roots of g with respect to t . In analytical level, Weyl group W ( G , T ) is defined as the quotient of normalizer by centralizer W ( G , T ) = N G ( T ) / Z G ( T ) . In algebraical level, Weyl group W ( Δ ( g , t ) ) is defined as the subgroup of the orthogonal group on g generated by the root reflections s α for α Δ ( g , t ) . When W ( G , T ) is considered as acting on ( i t 0 ) * , W ( G , T ) coincides with W ( Δ ( g , t ) ) [14].
In the case of twisted Weyl groups, let G be a compact connected Lie group with an Z n module structure, T be a maximal torus of G Z n , and g , t , g 0 and t 0 described as above. In [3], An J. defined the twisted Weyl group
W ( G , Z n , T ) : = W ( G , δ , T ) = N δ ( T ) / Z δ ( T ) ,
where δ is a generator of Z n . Motivated by the underlying equality between W ( G , T ) and W ( Δ ( g , t ) ) , one can also consider the algebraic twisted Weyl group. Since δ can be regarded as an automorphism on G, its differential d δ can be thought as an automorphism on g . Define
g α δ : = { X g a d ( d δ ) ( H ) X = α ( H ) X , H t } .
If g α δ 0 , α is called a twisted root of g with respect to t . Denote by Δ ( g , δ , t ) the set of twisted roots of g with respect to t . For α Δ ( g , δ , t ) , the twisted root reflection is defined as
s α δ ( γ ) : = γ 2 < d δ ( γ ) , α > α 2 α .
Define W ( Δ ( g , δ , t ) ) as the group generated by the twisted root reflections s α δ for α Δ ( g , δ , t ) . Similar as the proof in ([14], Theorem 4.54), one can obtain that
W ( G , Z n , T ) = W ( Δ ( g , δ , t ) ) ,
when W ( G , Z n , T ) is considered as acting on ( i t 0 ) * . For analytical twisted Weyl group W ( G , S 1 , T ) , we have an analogous algebraical counterpart described as above.
The twisted conjugate action τ δ of G associated to δ naturally induces the action of twisted Weyl group W ( S 1 ) on T :
W ( S 1 ) × T T , w · t : = τ g δ ( t ) ,
where δ is a topological generator of S 1 , w W ( S 1 ) , t T , g w . Now, we show the following property of the action of W ( S 1 ) on T.
Proposition 5.
Let G be a compact connected Lie group associated to an S 1 -module, δ be a topological generator of S 1 , T be a maximal compact torus of G S 1 and W ( S 1 ) be the twisted Weyl group of G associated to S 1 -module. Then two elements of T are δ-conjugate if and only if they are in one W ( S 1 ) -orbit.
Proof. 
Denote by O t W the W ( S 1 ) -orbit over t T , i.e.,
O t W : = { τ g δ ( t ) w W ( S 1 ) , g w } .
⇒) If t 1 , t 2 O t W , then there exist g 1 w 1 W ( S 1 ) , g 2 w 2 W ( S 1 ) such that t 1 = τ g 1 δ ( t ) , t 2 = τ g 2 δ ( t ) . Hence, t = g 1 1 t 1 δ ( g 1 ) . So,
t 2 = g 2 t δ ( g 2 ) 1 = g 2 g 1 1 t 1 δ ( g 1 ) δ ( g 2 ) 1 = ( g 2 g 1 1 ) t 1 δ ( g 2 g 1 1 ) 1 ,
which shows that t 2 = τ g 2 g 1 1 δ ( t 1 ) , g 2 g 1 1 G . So, t 1 , t 2 are δ -conjugate.
⇐) If t 1 , t 2 are δ -conjugate. i.e., there exists an element g G such that t 2 = τ g δ ( t 1 ) . It needs to show that t 1 , t 2 are in one W ( S 1 ) -orbit, i.e., to find an element w 0 W ( S 1 ) such that t 2 = w 0 · t 1 or to find an element g 0 N δ , G ( T ) such that
t 2 = τ g 0 δ ( t 1 ) .
Define
G t 1 : = { h G τ h δ ( t 1 ) = t 1 } .
Then T G t 1 , which holds for the fact T is connected and τ t δ ( t 1 ) = t 1 for all t T . Again,
h G t 1 τ h δ ( t 1 ) = h t 1 δ ( h ) 1 = t 1 h = t 1 δ ( h ) t 1 1 h G ı t 1 δ ,
where ı t 1 represents the automorphism induced by t 1 . Thus,
G t 1 = G ı t 1 δ .
Then by ([2], Theorem 2.1),
r a n k ( G δ ) = r a n k ( G ı t 1 δ ) = r a n k ( G t 1 ) ,
where r a n k ( G δ ) represents the dimension of the maximal compact torus of G δ . Hence, T is maximal compact torus of G t 1 . In fact, for the underlying g G , g 1 T g is also a maximal compact torus of G t 1 : for any t T , we have
τ g 1 t g δ ( t 1 ) = g 1 t g t 1 δ ( g 1 t g ) 1 = g 1 t g t 1 δ ( g ) 1 δ ( t ) 1 δ ( g 1 ) 1 = g 1 t t 2 t 1 δ ( g 1 ) 1 = g 1 t 2 δ ( g 1 ) 1 = t 1 ,
which shows that g 1 T g G t 1 . Thus, T and g 1 T g are both torus of G t 1 . Then there exists an element x G t 1 such that x T x 1 = g 1 T g . Pick g 0 = g x , then g 0 T g 0 1 = T .
Now we show g 0 N δ , G ( T ) and τ g 0 δ ( t 1 ) = t 2 . By
τ g 0 δ ( T ) = g 0 T δ ( g 0 ) 1 = g 0 T g 0 1 g 0 t 1 δ ( g 0 ) 1 = T g x t 1 δ ( x ) 1 δ ( g ) 1 = T g t 1 δ ( g ) 1 = T t 2 = T ,
we get g 0 N δ , G ( T ) ; by
τ g 0 δ ( t 1 ) = g 0 t 1 δ ( g 0 ) 1 = g x t 1 δ ( g x ) 1 = g x t 1 δ ( x ) 1 δ ( g ) 1 = g t 1 δ ( g ) 1 = t 2 ,
we get τ g 0 δ ( t 1 ) = t 2 . □

3. Twisted Weyl Groups of Compact Lie Groups and Nonabelian Cohomology

In this section, we discuss the relationship between twisted Weyl group W ( S 1 ) and the first nonabelian cohomology H 1 ( S 1 , G ) of S 1 with coefficients in G.
Let T be a maximal compact torus of G S 1 . As T is abelian and S 1 acts trivially on T, H 1 ( S 1 , T ) coincides with Z 1 ( S 1 , T ) and γ Z 1 ( S 1 , T ) is a group homomorphism γ : S 1 T . Thus, W ( S 1 ) naturally acts on H 1 ( S 1 , T ) = Z 1 ( S 1 , T ) , i.e.,
W ( S 1 ) × H 1 ( S 1 , T ) H 1 ( S 1 , T ) ,
( w · γ ) ( a ) : = g γ ( a ) a ( g ) 1 ,
where w W ( S 1 ) , γ H 1 ( S 1 , T ) , g w , a S 1 . Now, we show that the definition is well-defined. It suffices to show that w · γ Z 1 ( S 1 , T ) .
In fact, for any a 1 , a 2 S 1 , we have
( w · γ ) ( a 1 ) a 1 ( ( w · γ ) ( a 2 ) ) = g γ ( a 1 ) a 1 ( g ) 1 a 1 ( g γ ( a 2 ) a 2 ( g ) 1 ) = g γ ( a 1 ) a 1 ( g ) 1 a 1 ( g ) a 1 ( γ ( a 2 ) ) a 1 ( a 2 ( g ) 1 ) = g γ ( a 1 ) γ ( a 2 ) ( a 1 a 2 ) ( g ) 1 = g γ ( a 1 a 2 ) ( a 1 a 2 ) ( g ) 1 = ( w · γ ) ( a 1 a 2 ) ,
which shows that the underlying definition is well-defined. Thus, we have
( w · γ ) ( a 1 a 2 ) = ( w · γ ) ( a 1 ) ( w · γ ) ( a 2 ) .
Hence, w · γ is a group homomorphism.
The natural embedding
i : T G
induces the natural map
i 1 : H 1 ( S 1 , T ) H 1 ( S 1 , G ) , [ γ ] [ i γ ] ,
and i 1 can be reduced to
i 1 W ( S 1 ) : W ( S 1 ) \ H 1 ( S 1 , T ) H 1 ( S 1 , G ) , O γ W [ i γ ] ,
where O γ W : = { w · γ w W } is the W ( S 1 ) -orbit over γ in H 1 ( S 1 , T ) . Thus, we construct a correspondence between W ( S 1 ) \ H 1 ( S 1 , T ) and H 1 ( S 1 , G ) by i 1 W ( S 1 ) .
Theorem 1.
Let G be a compact connected Lie group associated to an S 1 -module, δ be a topological generator of S 1 , T be a maximal compact torus of G S 1 and W ( S 1 ) be the twisted Weyl group of G associated to S 1 -module. Then the map
i 1 W ( S 1 ) : W ( S 1 ) \ H 1 ( S 1 , T ) H 1 ( S 1 , G )
is a bijection.
Proof. 
By ([17], Theorem 2.5), the natural map i 1 is a surjection and hence i 1 W ( S 1 ) is also a surjection.
Now we show i 1 W ( S 1 ) is an injection. Suppose that γ 1 , γ 2 H 1 ( S 1 , T ) = Z 1 ( S 1 , T ) have the same image under i 1 . Then, there exists an element g G such that
γ 2 ( a ) = g γ ( a ) a ( g ) 1 ,
for all a A . Picking a topological generator δ 0 of S 1 . Thus, we have
γ 2 ( δ 0 ) = g γ 1 ( δ 0 ) δ 0 ( g ) 1 ,
which shows that γ 1 ( δ 0 ) and γ 2 ( δ 0 ) are δ 0 cojugate. By Proposition 5, γ 1 ( δ 0 ) and γ 2 ( δ 0 ) are in one W ( S 1 ) = W ( δ 0 ) -orbit. In other words, there exists an element g 0 N δ 0 , G ( T ) such that
γ 2 ( δ 0 ) = g 0 γ 1 ( δ 0 ) δ 0 ( g 0 ) 1 ,
which also means that there exists w 0 W such that g 0 w 0 and
γ 2 ( δ 0 ) = ( w 0 · γ 1 ) ( δ 0 ) .
Since δ 0 is a topological generator of S 1 , one can obtain that
γ 2 ( a ) = ( w 0 · γ 1 ) ( a ) ,
for all a S 1 . Thus,
γ 2 = w 0 · γ 1 ,
which shows that γ 1 and γ 2 are in one W ( S 1 ) -orbit. So,
i 1 W ( S 1 ) ( O γ 1 W ) = i 1 W ( S 1 ) ( O γ 2 W ) ,
and thus,
O γ 1 S 1 = O γ 2 S 1 ,
which shows that i 1 W ( S 1 ) is an injection. □
Remark 2.
Let A be a general compact group [4] and G be a compact connected Lie group with an A-module structure. Our motivation for this paper is to define the twisted Weyl group of G associated A-module. For this aim, we have to deal with the existence of the maximal compact torus in G 0 A , where G 0 A is the the identity connected component of
G A : = { g G a ( g ) = g , a A } .
However, for a general compact group A, maximal compact torus in G 0 A may not exist [1]. Even for the existence of A invariant maximal compact torus in G, it is not certain too [17]. From the discussions for the existence of maximal compact torus in G 0 A and G A , we find the following two open problems.
Problem 1.
Under what conditions there exists a nontrivial maximal compact torus in G 0 A ?
Problem 2.
If T is a nontrivial maximal compact torus of G 0 A . Can the maximal compact torus T generalize to an A invariant maximal compact torus T of G?

4. Discussions

In [9], Bartholdi studied the amenability of Γ -set, here Γ is a group, which was induced by John Von Neumann in 1929. Fundamentally, the notion exhibited the following property of a group acting on a Γ -set X: The Γ -set X right is called amenable if there exists a Γ -invariant mean m on the power set 2 X of X, namely a function
m : 2 X [ 0 , 1 ] ,
satisfying m ( A B ) = m ( A ) + m ( B ) , m ( X ) = 1 and
m ( A g ) = m ( A ) ,
for all A X and g Γ .
In [9], Bartholdi presented some criterions to show a Γ -set X amenable. By ([9], Proposition 2.3), one can get that the underlying twisted Weyl group W ( S 1 ) is indeed amenable. For the amenability of twisted Weyl groups, we will study it in a sole paper.

Author Contributions

Conceptualization, M.L. and X.Z.; methodology, M.L.; software, X.Z.; validation, M.L. and X.Z.; formal analysis, M.L.; investigation, M.L.; resources, X.Z.; data curation, M.L.; writing—original draft preparation, M.L.; writing—review and editing, X.Z.; visualization, M.L.; supervision, X.Z.; project administration, X.Z.; funding acquisition, X.Z. All authors have read and agreed to the published version of the manuscript.

Funding

This work research was funded by the National Natural Science Foundation of Tianjin (Grant No. 18JCYBJC18900), the National Natural Science Foundation of China (Grant No. 11301380)and the Higher School Science and Technology Development Fund Project in Tianjin (Grant No. 20131003).

Acknowledgments

The author is grateful to Jinpeng An for his invaluable directions. During the writing of this paper, the author benefits a lot from many help discussions with him. The authors would like to thank the referees for their many invaluable suggestions.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Bourbaki, N. Lie Groups and Lie Algebras: Chapters 7–9; Sringer: Berlin, Germany, 2005. [Google Scholar]
  2. An, J.P.; Wang, Z.D. Nonabelian cohomologies with coefficients in Lie groups. Trans. Am. Math. Soc. 2008, 360, 3019–3040. [Google Scholar] [CrossRef] [Green Version]
  3. An, J.P. Twisted Weyl groups of Lie groups and nonabelian cohomology. Geom. Dedicata 2007, 128, 167–176. [Google Scholar] [CrossRef] [Green Version]
  4. Hofmann, K.; Morris, S. The Structure of Compact Groups; Walter de Gruyter: Berlin, Germany, 1998. [Google Scholar]
  5. An, J.P.; Liu, M.; Wang, Z.D. Nonabelian cohomology of compact Lie groups. J. Lie Theory 2009, 19, 231–236. [Google Scholar]
  6. Guariglia, E. Riemann zeta fractional derivative-functional equation and link with primes. Adv. Differ. Equ. 2019, 2019, 261. [Google Scholar] [CrossRef]
  7. Guariglia, E. Fractional Derivative of the Riemann Zeta Function. In Fractional Dynamics; Cattani, C., Srivastava, H., Yang, X.J., Eds.; De Gruyter Open: Berlin, Germany, 2015; pp. 357–368. [Google Scholar]
  8. Guariglia, E.; Silvestrov, S. Fractional-Wavelet Analysis of Positive definite Distributions and Wavelets on D’(C). In Engineering Mathematics II; Silvestrov, S., Rancic, M., Eds.; Springer: Cham, Switzerland, 2016; pp. 337–353. [Google Scholar]
  9. Bartholdi, L. Amenability of Groups and G-Sets. In Sequences, Groups, and Number Theory; Berthé, V., Rigo, M., Eds.; Trends in Mathematics; Birkhäuser: Cham, Switzerland, 2018; pp. 433–544. [Google Scholar]
  10. Rahimi, H.; Nabizadeh, K. Amenability Modulo an ideal of Second duals of semigroup Algebras. Mathematics 2016, 4, 55. [Google Scholar] [CrossRef] [Green Version]
  11. Xin, Q.L.; Jiang, L.N.; Ma, Z.H. The basic construction from the conditional expectation on the quantum double of a finite group. Czechoslovak Math. J. 2015, 65, 347–359. [Google Scholar] [CrossRef] [Green Version]
  12. Shen, C.C.; Jiang, L.N.; Ma, Z.H. C*-algebra-valued G-Metric spaces and Related Fixed-Point Theorem. J. Funct. Spaces 2018, 2018, 3257189. [Google Scholar] [CrossRef] [Green Version]
  13. Balnojan, S.; Hertling, C. Reduced and Nonreduced Presentations of Weyl group Elements. J. Lie Theory 2019, 29, 559–599. [Google Scholar]
  14. Knapp, A.W. Lie Groups Beyond an Introduction, 2nd ed.; Birkhäuser: Boston, MA, USA, 2002. [Google Scholar]
  15. Serre, J.P. Galois Cohomology; Springer: Berlin, Germany, 1997. [Google Scholar]
  16. Weil, A. Remarks on the cohomology of groups. Ann. Math. 1964, 80, 149–157. [Google Scholar] [CrossRef] [Green Version]
  17. Liu, M. Nonabelian Cohomlogy of Compact Lie Groups. Ph.D. Thesis, PeKing University, Beijing, China, 2010. [Google Scholar]

Share and Cite

MDPI and ACS Style

Liu, M.; Zhang, X. Twisted Weyl Groups of Compact Lie Groups and Nonabelian Cohomology. Mathematics 2020, 8, 21. https://doi.org/10.3390/math8010021

AMA Style

Liu M, Zhang X. Twisted Weyl Groups of Compact Lie Groups and Nonabelian Cohomology. Mathematics. 2020; 8(1):21. https://doi.org/10.3390/math8010021

Chicago/Turabian Style

Liu, Ming, and Xia Zhang. 2020. "Twisted Weyl Groups of Compact Lie Groups and Nonabelian Cohomology" Mathematics 8, no. 1: 21. https://doi.org/10.3390/math8010021

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop