From relative Rota-Baxter operators and relative averaging operators on Lie algebras to relative Rota-Baxter operators on Leibniz algebras: a uniform approach
Abstract.
In this paper, first we construct two subcategories (using symmetric representations and antisymmetric representations) of the category of relative Rota-Baxter operators on Leibniz algebras, and establish the relations with the categories of relative Rota-Baxter operators and relative averaging operators on Lie algebras. Then we show that there is a short exact sequence describing the relation between the controlling algebra of relative Rota-Baxter operators on a Leibniz algebra with respect to a symmetric (resp. antisymmetric) representation and the controlling algebra of the induced relative Rota-Baxter operators (resp. averaging operators) on the canonical Lie algebra associated to the Leibniz algebra. Finally, we show that there is a long exact sequence describing the relation between the cohomology groups of a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric (resp. antisymmetric) representation and the cohomology groups of the induced relative Rota-Baxter operator (resp. averaging operator) on the canonical Lie algebra.
Key words and phrases:
Rota-Baxter operator, averaging operator, cohomology, Lie algebra, Leibniz algebra1. Introduction
The purpose of this paper is to provide a unified approach to study relative Rota-Baxter operators and relative averaging operators on Lie algebras.
1.1. Relative Rota-Baxter operators on Lie algebras
G. Baxter introduced the concept of a Rota-Baxter commutative algebra [8] in his study of fluctuation theory in probability. A Rota-Baxter operator of weight zero on an associative algebra is a linear map such that
These operators with general weights have been found many applications in recent years, including the algebraic approach of Connes-Kreimer [13] to renormalization of perturbative quantum field theory, quantum analogue of Poisson geometry [48], twisting on associative algebras [49], dendriform algebras and associative Yang-Baxter equations [1]. The notion of Rota-Baxter operators on Lie algebras was introduced independently in the 1980s as the operator form of the classical Yang-Baxter equation. A linear operator on a Lie algebra is called a Rota-Baxter operator of weight zero if the following condition is satisfied:
Moreover, Kupershmidt introduced the notion of a relative Rota-Baxter operator (also called an -operator) on a Lie algebra with respect to arbitrary representation in [29]. Note that a skew-symmetric classical -matrix is a relative Rota-Baxter operator on a Lie algebra with respect to the coadjoint representation. Relative Rota-Baxter operators play important roles in the study of integrable systems [6, 40], provide solutions of the classical Yang-Baxter equation in the semidirect product Lie algebra and give rise to pre-Lie algebras [5]. See [21] for more details. Recently, the deformation and homotopy theory of relative Rota-Baxter Lie algebras and relative Rota-Baxter associative algebras were established in [15, 30, 44, 45].
1.2. Averaging operators on Lie algebras
An averaging operator on an associative algebra is a linear operator such that
In the last century, many studies on averaging operators were done for various special algebras, such as function spaces, Banach algebras, and the topics and methods were largely analytic [7, 12, 24, 39]. Recently, it was found that commutative double algebra structures [20] on a vector space can be described by symmetric averaging operators on the associative algebra . Averaging operators can be defined on algebras over arbitrary binary operad [1]. In particular, a linear operator on a Lie algebra is called an averaging operator if the following condition is satisfied:
There is a close relation between the classical Yang-Baxter equation, conformal algebras and averaging operators on Lie algebras [26].
Averaging operators on Lie algebras are also called embedding tensors in some mathematical physics literatures. Embedding tensors and their associated tensor hierarchies provide an algebraic and efficient way to construct supergravity theories and further to construct higher gauge theories (see e.g. [10, 23, 28, 42]). Recently the controlling algebra and the cohomology theory of embedding tensors were established in [41] using the higher derived bracket. Meanwhile the cohomology and homotopy theory of averaging associative algebras were studied in [52].
The concept of Rota-Baxter operators and averaging operators were further studied on the level of algebraic operads in [3, 35, 36, 37, 50, 51]. The action of the Rota-Baxter operator on a binary quadratic operad splits the operad, and the action of the averaging operator on a binary quadratic operad duplicates the operad. Since the splitting and duplication processes are in Koszul duality, one can regard the Rota-Baxter operator and averaging operator to be Koszul dual to each other.
1.3. Relative Rota-Baxter operators on Leibniz algebras
Leibniz algebras were first discovered by Bloh who called them D-algebras [9]. Then Loday rediscovered this algebraic structure and called them Leibniz algebras with the motivation in the study of the periodicity in algebraic K-theory [31, 33, 34]. Averaging operators on Lie algebras give rise to Leibniz algebras, which can be viewed as the duplication of Lie algebras. From the viewpoint of the operad theory, leibniz algebras are defined as the algebras over the operad which is the duplicator of the operad [36]. More intrinsically, let be a binary quadratic operad. The algebra duplicates by an averaging operator on the -algebra is a -algebra. On the other hand, the algebra splits by a Rota-Baxter operator on the -algebra is a -algebra. Since is the Koszul dual of [36], the duplication construction of an averaging operator is a kind of Koszul dual of the splitting construction of a Rota-Baxter operator [3, 36]. The (co)homology and homotopy theories of Leibniz algebras were established in [18, 33, 38, 2]. Recently Leibniz algebras were studied from different aspects due to applications in both mathematics and physics. In particular, integration of Leibniz algebras were studied in [11, 14] and deformation quantization of Leibniz algebras was studied in [16]. As the underlying algebraic structures of embedding tensors and Courant algebroids, Leibniz algebras also have application in higher gauge theories [10, 28, 43] and homogeneous spaces [25]. The notion of relative Rota-Baxter operators on Leibniz algebras was introduced in [46] to study Leibniz bialgebras. Moreover, the cohomology theory of relative Rota-Baxter operators on Leibniz algebras was given in [47] and used to classify linear deformations and formal deformations.
1.4. Outline of the paper
In this paper, we propose a unified approach to study relative Rota-Baxter operators and relative averaging operators on Lie algebras using relative Rota-Baxter operators on Leibniz algebras. First we observe that there are forgetful functors from the categories of relative Rota-Baxter operators and relative averaging operators on Lie algebras to the category of relative Rota-Baxter operators on Leibniz algebras. Conversely, we construct two functors from certain subcategories of the category of relative Rota-Baxter operators on Leibniz algebras to the categories of relative Rota-Baxter operators and relative averaging operators on Lie algebras, and show that they are left adjoint for the aforementioned forgetful functors. Then we show that the controlling algebras of relative Rota-Baxter operators and relative averaging operators on Lie algebras can be derived from the controlling algebra of relative Rota-Baxter operators on Leibniz algebras. Finally, we give a long exact sequence to describe the relation between the cohomology groups of a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric (resp. antisymmetric) representation and the cohomology groups of the induced relative Rota-Baxter operator (resp. averaging operator) on the canonical Lie algebra. The concrete relations can be summarized by the following diagrams:
The paper is organized as follows. In Section 2, using symmetric representations and antisymmetric representations, we construct two subcategories of the category of relative Rota-Baxter operators on Leibniz algebras, and establish the relations with the categories of relative Rota-Baxter operators and relative averaging operators on Lie algebras. In Section 3, we show that there is a short exact sequence describing the relation between the controlling algebra of relative Rota-Baxter operators on a Leibniz algebra with respect to a symmetric representation and the controlling algebra of the induced relative Rota-Baxter operators on the canonical Lie algebra. For relative averaging operators, we establish a similar result. In Section 4, first we introduce the Loday-Pirashvili cohomology of a relative Rota-Baxter operator on a Lie algebra and then use a long exact sequence to describe the relation between the cohomology groups of a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation and the Loday-Pirashvili cohomology groups of the induced relative Rota-Baxter operator on the canonical Lie algebra. Similarly, there is a long exact sequence describing the relation between the cohomology groups of a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation and the cohomology groups of the induced relative averaging operator on the canonical Lie algebra.
In this paper, we work over an algebraically closed field of characteristic 0 and all the vector spaces are over and finite-dimensional.
2. Relations between the categories , and
In this section, we establish the relations between certain subcategories of the category of relative Rota-Baxter operators on Leibniz algebras and the category of relative Rota-Baxter operators on Lie algebras as well as the category of relative averaging operators on Lie algebras.
2.1. Relative Rota-Baxter operators and relative averaging operators
Definition 2.1.
([29]) A relative Rota-Baxter operator on a Lie algebra with respect to a representation is a linear map satisfying the following quadratic constraint:
(1) |
Let (resp. ) be a relative Rota-Baxter operator on the Lie algebra (resp. ) with respect to the representation (resp. ). A homomorphism from to is a pair , where is a Lie algebra homomorphism, is a linear map such that
(2) | |||||
(3) |
In particular, if and are invertible, then is called an isomorphism.
We denote by the category of relative Rota-Baxter operators on Lie algebras.
Definition 2.2.
([1, 39]) A relative averaging operator on a Lie algebra with respect to a representation is a linear map satisfying the following quadratic constraint:
(4) |
Let (resp. ) be a relative averaging operator on the Lie algebra (resp. ) with respect to the representation (resp. ). A homomorphism from to is a pair , where is a Lie algebra homomorphism, is a linear map such that
(5) | |||||
(6) |
In particular, if and are invertible, then is called an isomorphism.
We denote by the category of relative averaging operators on Lie algebras.
A (left) Leibniz algebra is a vector space together with a bilinear operation such that
A representation of a Leibniz algebra is a triple , where is a vector space, are linear maps such that for all ,
Here is the commutator Lie bracket on . There are two important kinds of representations of Leibniz algebras:
-
•
a representation is called symmetric if ;
-
•
a representation is called antisymmetric if .
For any representation of a Leibniz algebra , we define a vector subspace which is spanned by all elements with and . Directly from for all , we deduce that is an antisymmetric representation. The quotient representation is then symmetric, denoted by , and we have an exact sequence
(7) |
Now we recall the notion of a relative Rota-Baxter operator on a Leibniz algebra.
Definition 2.3.
([46]) Let be a representation of a Leibniz algebra . A linear operator is called a relative Rota-Baxter operator on with respect to if satisfies:
(8) |
Let (resp. ) be a relative Rota-Baxter operator on the Leibniz algebra (resp. ) with respect to the representation (resp. ). A homomorphism from to is a pair , where is a Leibniz algebra homomorphism, is a linear map such that for all ,
(9) | |||||
(10) | |||||
(11) |
In particular, if and are invertible, then is called an isomorphism from to .
We denote by the category of relative Rota-Baxter operators on Leibniz algebras. Moreover, we denote by and the categories of relative Rota-Baxter operators on Leibniz algebras with respect to symmetric representations and antisymmetric representations respectively.
2.2. Relation to crossed modules
Note that relative Rota-Baxter operators on Lie algebras and Leibniz algebras resp. as well as relative averaging operators on Lie algebras give rise to a bracket on .
For a relative Rota-Baxter operator on a Lie algebra , the bracket
for all renders a Lie algebra such that becomes a morphism of Lie algebras.
In the same vein, given a relative Rota-Baxter operator on a Leibniz algebra , the bracket
for all renders a Leibniz algebra such that becomes a morphism of Leibniz algebras.
Still in the same vein, given a relative averaging operator on a Lie algebra , the bracket
for all renders a Leibniz algebra such that is a morphism of Leibniz algebras.
The next proposition shows that with an equivariance condition, the action of (resp. ) on V becomes an action by derivations with respect to these brackets. This kind of equivariance condition is present, for example, in Loday-Pirashvili’s tensor category of linear maps, see [32].
Proposition 2.4.
-
(1)
Let be a relative Rota-Baxter operator on a Lie algebra and suppose that is equivariant in the sense that for all and all . We deduce that for all is a derivation of the Lie algebra . Thus, we obtain that is homomorphism of Lie algebras.
-
(2)
Let be a relative Rota-Baxter operator on a Leibniz algebra and suppose that is equivariant in the sense that and for all and all . Then acts on by derivations (in the Leibniz sense), i.e. for all and all
We deduce that the pair for all is a biderivation [31] of the Leibniz algebra . For relative averaging operators, equivariance will imply that the left action is by derivations, but defining the right operation to be the opposite of the left action or zero will not render this an action by derivations in the Leibniz sense in general.
Note that this proposition does not give rise to crossed modules, because the requirements in the Lie case and in the Leibniz case for are not fullfilled. In the case of relative averaging operators, this requirement is fullfilled, but the action is not by derivations in general. On the other hand, it is well known that crossed modules of Lie algebras provide examples of relative averaging operators, see Example 2.14 in [41].
2.3. Relations between the categories and
First we construct a functor from the category of relative Rota-Baxter operators on Lie algebras to the category of relative Rota-Baxter operators on Leibniz algebras with respect to symmetric representations.
On objects, the functor is defined as follows. Let be a relative Rota-Baxter operator on the Lie algebra with respect to the representation . We view the Lie algebra as a Leibniz algebra. Moreover the Lie algebra representation gives rise to a symmetric representation of the Leibniz algebra . Then the equality (1) means that is a relative Rota-Baxter operator on the Leibniz algebra with respect to the symmetric representation .
On morphisms, the functor is defined as follows. Let (resp. ) be a relative Rota-Baxter operator on the Lie algebra (resp. ) with respect to the representation (resp. ). Let be a homomorphism from the relative Rota-Baxter operator to . Then it is obvious that is also a homomorphism between the above induced relative Rota-Baxter operators on Leibniz algebras.
It is straightforward to see that defined above is indeed a functor.
In the sequel, we construct a functor from the category to the category .
Let be a Leibniz algebra. Denote by the ideal of squares spanned by all elements for all . We call the Leibniz kernel of . Observe that for the adjoint representation with and . The following result is obvious.
Lemma 2.5.
Let be a Leibniz algebra. Then
is naturally a Lie algebra in which we denote the Lie bracket by .
We call the canonical Lie algebra associated to the Leibniz algebra . Observe that for the adjoint representation . The equivalence class of in will be denoted by . Moreover, is contained in the left center of the Leibniz algebra We denote by
(12) |
the natural projection from the Leibniz algebra to the canonical Lie algebra . Obviously, the map preserves the bracket operation, i.e. the following equality holds:
(13) |
Let be a representation of a Leibniz algebra . Then is a Leibniz algebra homomorphism. We deduce that
(14) |
Thus, there is exactly one Lie algebra homomorphism which is defined by
(15) |
such that following diagram of Leibniz algebra homomorphisms commute:
It implies that is a representation of the canonical Lie algebra
Proposition 2.6.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . Then is a relative Rota-Baxter operator on the canonical Lie algebra with respect to the representation .
Proof.
Applying the projection to the both sides of (8), we obtain
which implies that is a relative Rota-Baxter operator on the canonical Lie algebra with respect to the representation . ∎
Proposition 2.7.
Let (resp. ) be a relative Rota-Baxter operator on the Leibniz algebra (resp. ) with respect to the symmetric representation (resp. ). Let be a homomorphism of relative Rota-Baxter operators on Leibniz algebras from to . Then is a homomorphism of relative Rota-Baxter operators on Lie algebras from to , where the Lie algebra homomorphism is defined by
Proof.
By , we obtain that induces the Lie algebra homomorphism Then it is straightforward to deduce that is a homomorphism of relative averaging operators from to . ∎
Now we are ready to give the main result in this subsection.
Theorem 2.8.
There is a functor from the category of relative Rota-Baxter operators on Leibniz algebras with respect to symmetric representations to the category of relative Rota-Baxter operators on Lie algebras, such that is a left adjoint for .
Proof.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . By Proposition 2.6, is a relative Rota-Baxter operator on the canonical Lie algebra with respect to the representation . Thus, on objects, we define
Let be a homomorphism of relative Rota-Baxter operators on Leibniz algebras from to . By Proposition 2.7, is a homomorphism of relative Rota-Baxter operators from to . Thus, on morphisms, we define
Then it is straightforward to see that is indeed a functor. Moreover, let be a Lie algebra, we have . Thus, we obtain that Thus, we have the identity natural transformation (the counit of the adjunction)
Moreover, for any relative Rota-Baxter operator on a Lie algebra with respect to a representation , relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation and relative Rota-Baxter operator homomorphism from to , we deduce that is a Leibniz algebra homomorphism from to and
Thus, there is exactly one relative Rota-Baxter operator homomorphism from to such that the following diagram commutes:
which implies that is a left adjoint for . The proof is finished. ∎
Remark 2.9.
The functors and are induced by those establishing the equivalence of categories between modules over a Lie algebra and symmetric Leibniz representations over , viewed as a Leibniz algebra. From this, one may deduce that the pair constitutes almost an adjoint equivalence. In fact, is fully faithful, because the counit of the adjunction is a natural isomorphism for each object of . On the other hand, the unit of the adjunction is not an isomorphism. For an object in , corresponds to viewing again as an object in . We have a natural relative Rota-Baxter operator homomorphism from to
but it is not an isomorphism in general.
Note that one may add the data which arises from the passage from to to the category in order to make it an equivalence of categories. The data consists of the isomorphism class of an antisymmetric representation of (corresponding to ) and of the Loday-Pirashvili cohomology class of a -cocycle characterizing the abelian extension
(16) |
Given a relative Rota-Baxter operator together with a -module (viewed as an antisymmetric representation of the Leibniz algebra ) and a -cocycle , we obtain a relative Rota-Baxter operator for the Leibniz algebra constructed from with and viewed as a symmetric representation of , because acts trivially from the left and thus also trivially from the right as the representation is symmetric. This construction depends on the choice of the cocycle and the module , but becomes natural when passing to cohomology and isomorphism classes.
2.4. Relations between the categories and
First we construct a functor from the category of relative averaging operators on Lie algebras to the category of relative Rota-Baxter operators on Leibniz algebras with respect to antisymmetric representations.
On objects, the functor is defined as follows. Let be a relative averaging operator on a Lie algebra with respect to a representation . We view the Lie algebra as a Leibniz algebra. Moreover the Lie algebra representation gives rise to an antisymmetric representation of the Leibniz algebra . Then the equality (4) means that is a relative Rota-Baxter operator on the Leibniz algebra with respect to the antisymmetric representation .
On morphisms, the functor is defined as follows. Let (resp. ) be a relative averaging operator on a Lie algebra (resp. ) with respect to a representation (resp. ). Let be a homomorphism from the relative averaging operator to . Then it is obvious that is also a homomorphism between the above induced relative Rota-Baxter operators.
It is straightforward to see that defined above is indeed a functor.
In the sequel, we construct a functor from the category to the category .
Proposition 2.10.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . Then is a relative averaging operator on the Lie algebra with respect to the representation .
Proof.
Applying the projection to the both sides of (8), we obtain
which implies that is an averaging operator on the Lie algebra with respect to the representation . ∎
Similar to Proposition 2.7, we have the following result.
Proposition 2.11.
Let (resp. ) be a relative Rota-Baxter operator on the Leibniz algebra (resp. ) with respect to the antisymmetric representation (resp. ). Let be a homomorphism of relative Rota-Baxter operators on Leibniz algebras from to . Then is a homomorphism of relative averaging operators from to , where the Lie algebra homomorphism is defined by
Now we are ready to give the main result in this subsection.
Theorem 2.12.
There is a functor from the category of relative Rota-Baxter operators on Leibniz algebras with respect to antisymmetric representations to the category of relative averaging operators on Lie algebras, such that is a left adjoint for .
Proof.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . By Proposition 2.10, is a relative averaging operator on the Lie algebra with respect to the representation . Thus, on objects, we define
Let be a homomorphism of relative Rota-Baxter operators on Leibniz algebras from to . By Proposition 2.11, is a homomorphism of relative averaging operators from to . Thus, on morphisms, we define
Then it is straightforward to see that is indeed a functor.
By the similar argument as Theorem 2.8, we can show that is a left adjoint for . We omit details. ∎
Remark 2.13.
Like in Remark 2.9, the pair of functors and form almost an adjoint equivalence.
3. Relations between the controlling algebras
In this section, we establish the relation between the controlling algebra of relative Rota-Baxter operators on a Leibniz algebra with respect to a symmetric (resp. antisymmetric) representation and the controlling algebra of the relative Rota-Baxter (resp. averaging) operators on the canonical Lie algebra.
A permutation is called an -shuffle if and . If or , we assume . The set of all -shuffles will be denoted by . The notion of an -shuffle and the set are defined analogously.
A degree element is called a Maurer-Cartan element of a differential graded Lie algebra if it satisfies the Maurer-Cartan equation: . The set of Maurer-Cartan elements in a dgla will be denoted by .
Let be a vector space. We consider the graded vector space . An element is defined to have degree . The Balavoine bracket on the graded vector space is given by:
(17) |
where is defined by
(18) |
and is defined by
It is well known that
Theorem 3.1.
([4, 19]) With the above notations, is a graded Lie algebra. Its Maurer-Cartan elements (as a differential graded Lie algebra with zero differential) are precisely the Leibniz algebra structures on .
3.1. The controlling algebra of relative Rota-Baxter operators on a Leibniz algebra
Let be a representation of a Leibniz algebra . Then there is a Leibniz algebra structure on given by
(19) |
This Leibniz algebra is called the semidirect product of and , and denoted by We denote the above semidirect product Leibniz multiplication by Consider the graded vector space
where an element is defined to be of degree .
Theorem 3.2.
([46]) With the above notations, is a graded Lie algebra, where the graded Lie bracket is given by the derived bracket as following:
More precisely, we have
Moreover, its Maurer-Cartan elements (where the differential is taken to be zero again) are relative Rota-Baxter operators on the Leibniz algebra with respect to the representation .
In fact, this graded Lie algebra construction is functorial. More precisely, we consider the category - of representations of the Leibniz algebra . Then, we have a contravariant functor from the category - to the category of graded Lie algebras.
Let and be two representations of the Leibniz algebra . A homomorphism from to is a linear map such that
(20) | |||||
(21) |
Let be a homomorphism from the representation to . Define a linear map by
(22) |
Proposition 3.3.
With the above notations, is a homomorphism from the graded Lie algebra to .
Moreover, we have the following theorem.
Theorem 3.4.
Proof.
Let be a representation of the Leibniz algebra . For the identity homomorphism , we have . Moreover, let and be two homomorphisms of representations of the Leibniz algebra . For and , we have
Thus, we deduce that is a contravariant functor. ∎
As a consequence, the short exact sequence (7)
associated to the -representation induces homomorphisms of graded Lie algebras
which do not necessarily form an exact sequence of graded Lie algebras. By passage to the corresponding Maurer-Cartan sets, we obtain a sequence of induced maps
(25) |
which describe (loosely) how a relative Rota-Baxter operator on the Leibniz algebra with respect to is assembled from a relative Rota-Baxter operator on the Lie algebra and a relative averaging operator on , according to Theorems 2.8 and 2.12, see also Corollaries 3.6 and 3.9.
The (not necessarily exact) sequence (25) is the effect in the module-variable of the standard short exact sequence. In order to investigate also the effect in the Leibniz-algebra-variable of the standard short exact sequence (26) (which will be the subject of Theorem 3.5), we need some more preparation.
3.2. The controlling algebra of relative Rota-Baxter operators on a Lie algebra
Associated to any Leibniz algebra , we have an exact sequence of Leibniz algebras
(26) |
It is an abelian extension of by . Moreover the induced representation of on is an antisymmetric representation. Observe that the sequence (26) is the special case of the sequence (7) for the adjoint representation of .
Let be a representation of a Leibniz algebra . Thus, there is exactly one Lie algebra homomorphism , such that following diagram of Leibniz algebra homomorphisms commutes:
(31) |
Consider the category - of symmetric representations of the Leibniz algebra , which is a subcategory of -. Then we have the following functors
which are given by
-
•
the functor , which is defined on objects and on morphisms respectively by
(32) (33) -
•
the functor , which is defined on objects and on morphisms respectively by
(34) (35)
for symmetric representations and of the Leibniz algebra and .
Thus, we have three functors and from to , where the functor is given in Theorem 3.4. Moreover, for any symmetric representation of the Leibniz algebra , we define
and
as following:
(36) | |||||
(37) |
Theorem 3.5.
With the above notations, is a natural transformation from the functor to , and is a natural transformation from the functor to . Moreover, for any symmetric representation of the Leibniz algebra , we have the following short exact sequence of graded Lie algebras:
(38) |
Proof.
Since is an ideal of the Leibniz algebra , it follows that is a subalgebra of . Thus the linear embedding map is a graded Lie algebra homomorphism. Let be a homomorphism of symmetric representations of . It is straightforward to obtain the following commutative diagram:
which implies that is a natural transformation.
Since is a Leibniz algebra homomorphism and the definition (15) of , we deduce that is a homomorphism of graded Lie algebras. Moreover, let be a homomorphism of symmetric representations of and . Then we have
which implies that we have the following commutative diagram:
Thus, is a natural transformation.
Since is vector space over the field , for any positive integer , the functor is an exact functor from the category of vector spaces over to itself. Moreover, we have and . By the exact sequence of Leibniz algebras (26), we obtain the short exact sequence of graded Lie algebras (38). The proof is finished. ∎
Corollary 3.6.
Let be a symmetric representation of a Leibniz algebra . The Maurer-Cartan elements of the graded Lie algebra are exactly relative Rota-Baxter operators on the Lie algebra with respect to the representation given in (15).
Let be a representation of a Lie algebra . Then is a symmetric representation of the Leibniz algebra . By , we obtain a graded Lie algebra whose Maurer-Cartan elements are the relative Rota-Baxter operators on the Lie algebra with respect to the representation . On the other hand, in [44] the authors construct another graded Lie algebra on the graded vector subspace where the graded Lie bracket is given by
for all and . Its Maurer-Cartan elements are also relative Rota-Baxter operators on with respect to the representation .
Proposition 3.7.
The graded Lie algebra is a subalgebra of .
Proof.
Let be a representation of a Lie algebra . The semidirect product Lie algebra is the same as the semidirect product Leibniz algebra . We denote this semidirect product multiplication by Recall that the Nijenhuis-Richardson bracket associated to the direct sum vector space gives rise to a graded Lie algebra . The graded Lie algebra is also obtained via the derived bracket [27]:
Since is a graded Lie subalgebra of , we deduce that
which implies that is a subalgebra of . ∎
3.3. The controlling algebras of relative averaging operators on a Lie algebra
Denote by - the category of antisymmetric representations of the Leibniz algebra , which is a subcategory of -. Then we have the following functors
which are given by
-
•
the functor , which is defined on objects and on morphisms respectively by
(42) (43) -
•
the functor , which is defined on objects and on morphisms respectively by
(44) (45)
for antisymmetric representations and of the Leibniz algebra and .
Thus, we have three functors and from the category to the category . Moreover, for any antisymmetric representation of the Leibniz algebra , we define
and
as following:
Similar to Theorem 3.5, we have the following result.
Theorem 3.8.
With above notations, is a natural transformation from the functor to , and is a natural transformation from the functor to . Moreover, for any antisymmetric representation of the Leibniz algebra , we have the following short exact sequence of graded Lie algebras:
Proof.
The proof is parallel to that of Theorem 3.5, so we omit the details. ∎
Corollary 3.9.
Let be an antisymmetric representation of a Leibniz algebra . The Maurer-Cartan elements of the graded Lie algebra are exactly relative averaging operators on the Lie algebra with respect to the representation given in (15).
Let be a representation of a Lie algebra . Then is an antisymmetric representation of the Leibniz algebra . By , we obtain a graded Lie algebra whose Maurer-Cartan elements are the relative averaging operators on the Lie algebra with respect to the representation . This graded Lie algebra is exactly the same as the one given in [41].
4. Relations between the cohomologies
In this section, we establish the relations between the cohomology groups of a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric (resp. antisymmetric) representation and the cohomology groups of the induced relative Rota-Baxter (resp. averaging) operator on the canonical Lie algebra.
4.1. Cohomology of relative Rota-Baxter operators on Leibniz algebras
Definition 4.1.
([33]) Let be a representation of a Leibniz algebra . The Loday-Pirashvili cohomology of with coefficients in is the cohomology of the cochain complex , where and the coboundary operator is defined by
for all . The resulting cohomology is denoted by .
Theorem 4.2.
([47]) Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to . Define by
(46) |
Then is a Leibniz algebra. Moreover, define by
(47) |
Then is a representation of the Leibniz algebra .
Let be the corresponding Loday-Pirashvili coboundary operator of the Leibniz algebra with coefficients in the representation . More precisely, is given by
Definition 4.3.
([47]) Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a representation . The cohomology of the cochain complex
is taken to be the cohomology for the relative Rota-Baxter operator .
We denote the set of -cocycles by , the set of -coboundaries by and the -th cohomology group by
(48) |
Up to a sign, the coboundary operators coincides with the differential operator defined by using the Maurer-Cartan element .
Theorem 4.4.
([47]) Let be a relative Rota-Baxter operator on the Leibniz algebra with respect to the representation . Then we have
4.2. The Loday-Pirashvili cohomology of relative Rota-Baxter operators on Lie algebras
In [44] the authors established a cohomology theory of a relative Rota-Baxter operator on a Lie algebra as the Chevalley-Eilenberg cohomology of a Lie algebra. Thus we will refer to it as the Chevalley-Eilenberg cohomology of a relative Rota-Baxter operator on a Lie algebra. More precisely, let be a relative Rota-Baxter operator on a Lie algebra with respect to a representation . Define by
Then is a representation of the Lie algebra on the vector space . We denote by the Chevalley-Eilenberg cochain complex of the Lie algebra with coefficients in the representation , where and . The cohomology of the cochain complex is called the Chevalley-Eilenberg cohomology of the relative Rota-Baxter operator . Denote the set of -cocycles by and the set of -coboundaries by . Denote by
(49) |
the corresponding -th cohomology group.
As explained in Section 2, a relative Rota-Baxter operator on a Lie algebra with respect to a representation can be viewed as a special relative Rota-Baxter operator on the Leibniz algebra with respect to the symmetric representation . Thus, the above approach to define the cohomology of a relative Rota-Baxter operator on a Leibniz algebra can be applied to define a cohomology of a relative Rota-Baxter operator on a Lie algebra. More precisely, let be a relative Rota-Baxter operator on a Lie algebra with respect to a representation . Then is a Leibniz algebra, where the Leibniz bracket is defined by
Moreover, is a symmetric representation of the Leibniz algebra , where is defined by
Let be the corresponding Loday-Pirashvili coboundary operator of the Leibniz algebra with coefficients in the symmetric representation .
Definition 4.5.
Let be a relative Rota-Baxter operator on a Lie algebra with respect to a representation . The cohomology of the cochain complex is called the Loday-Pirashvili cohomology of the relative Rota-Baxter operator .
We denote the set of -cocycles by , the set of -coboundaries by and the -th cohomology group by
(50) |
Remark 4.6.
Note that the monomorphism induced from the natural epimorphism induces isomorphisms
and a long exact sequence
where is the cohomology of the relative complex which is by definition (up to a degree shift) the cokernel of the monomorphism . These matters as well as a spectral sequence linking and in still another way can be found in [18]. In particular, Theorem 2.6 in [18] shows how the vanishing of Chevalley-Eilenberg cohomology implies the vanishing of Loday-Pirashvili cohomology.
Remark 4.7.
Note furthermore that the comparison between Chevalley-Eilenberg cohomologies and Loday-Pirashvili cohomologies makes sense more generally for any relative Rota-Baxter operator over a Leibniz algebra as soon as the representation of on is symmetric, because then the bracket on is a Lie bracket.
In case the induced representation of on is not symmetric, one may use the short exact sequence
and the relation of Loday-Pirashvili cohomology with values in an antisymmetric representation to the cohomology with values in a symmetric representation (see [18]) in order to reduce the computations to the case of symmetric coefficients.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . By Proposition 2.6, is a relative Rota-Baxter operator on the Lie algebra with respect to the representation given by (15). Now we establish the relationship between the cohomology groups of the relative Rota-Baxter operator and the Loday-Pirashvili cohomology groups of the relative Rota-Baxter operator
Lemma 4.8.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . Then is a subrepresentation of .
Proof.
By the fact that is contained in the left center of the Leibniz algebra , we have
which implies that . Therefore, is a subrepresentation of . ∎
Note that even though the Leibniz algebra becomes a Lie algebra if is a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation, the representation given in Theorem 4.2 is still a representation of as a Leibniz algebra, and But the quotient representation is symmetric, where and are given by
(51) |
Proposition 4.9.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . Then the quotient representation is a symmetric representation of as a Leibniz algebra, i.e. .
Proof.
For all , we have
Thus, we deduce that . ∎
Let be the corresponding Loday-Pirashvili coboundary operator of the Leibniz algebra with coefficients in the quotient representation .
Proposition 4.10.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . Then
That is, the Loday-Pirashvili cochain complex associated to the relative Rota-Baxter operator and the cochain complex obtained by using the quotient representation are the same.
Proof.
Obviously, the spaces of -cochains are the same. Thus we only need to show that the induced Leibniz algebra structure on by the relative Rota-Baxter operator is the same as , and the representations and are the same.
For all , by (15), we have
For any , we have
Thus ∎
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . By Lemma 4.8, we have the following short exact sequence of representations of the Leibniz algebra :
(52) |
Moreover, we have the following result:
Theorem 4.11.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to a symmetric representation . Then there is a short exact sequence of the cochain complexes:
Consequently, there is a long exact sequence of the cohomology groups:
(53) |
where the connecting map is defined by for all
Proof.
For any and , we have
which implies that is a homomorphism of cochain complexes. For and , we have
For , by Theorem 4.4 and Theorem 3.5, for all , we have
Thus, we deduce that is a homomorphism of cochain complexes. Since is vector space over the field , for any positive integer , the functor is an exact functor from the category of vector spaces over to itself. Moreover, we have and . By the short exact sequence (52) of representations of the Leibniz algebra , we obtain the short exact sequence of the above cochain complexes. The proof is finished. ∎
Remark 4.12.
In the situation of Theorem 4.11, the short exact sequence of cochain complexes can be completed into a diagram
which leads then to a corresponding exact diagram in cohomology (starting from degree , while there are isomorphisms in degree and as before).
4.3. Cohomology of relative averaging operators on Lie algebras
As explained in Section 2, a relative averaging operator on a Lie algebra with respect to a representation can be viewed as a special relative Rota-Baxter operator on the Leibniz algebra with respect to the antisymmetric representation . Thus, the above approach to define the cohomology of relative Rota-Baxter operators on Leibniz algebras can be applied to define the cohomology of relative averaging operators. More precisely, let be a relative averaging operator on a Lie algebra with respect to a representation . Then is a Leibniz algebra, where the Leibniz bracket is defined by
Moreover, is a representation of the Leibniz algebra , where is defined by
Let be the corresponding Loday-Pirashvili coboundary operator of the Leibniz algebra with coefficients in the representation . More precisely, is given by
Definition 4.13.
Let be a relative averaging operator on a Lie algebra with respect to a representation . The cohomology of the cochain complex is defined to be the cohomology of the relative averaging operator .
We denote the set of -cocycles by , the set of -coboundaries by and the -th cohomology group by
(54) |
See [41] for more details about the cohomology theory of a relative averaging operator on a Lie algebra.
Let be a relative Rota-Baxter operator on the Leibniz algebra with respect to an antisymmetric representation . By Proposition 2.10, is a relative averaging operator on the Lie algebra with respect to the representation given by (15). Now we establish the relationship between the cohomology groups of the relative Rota-Baxter operator and the cohomology groups of the relative averaging operator
Similarly as Lemma 4.8, we have the following result.
Lemma 4.14.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . Then is a subrepresentation of .
By Lemma 4.14, we have the quotient representation , where and are given by (51). Let be the corresponding Loday-Pirashvili coboundary operator of the Leibniz algebra with coefficients in the quotient representation . Similar to Proposition 4.10, we have the following result.
Proposition 4.15.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . Then
That is, the cochain complex associated to the relative averaging operator and the cochain complex obtained by using the quotient representation are the same.
Proof.
Obviously, the spaces of -cochains are the same. Thus we only need to show that the induced Leibniz algebra structure on by the relative averaging operator is the same as , and the representations and are the same.
For all , by (15), we have
For any , we have
Thus, ∎
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . By Lemma 4.14, we have the following short exact sequence of representations of the Leibniz algebra :
(55) |
Similar to Theorem 4.11, we have the following result describing the relation among various cohomology groups.
Theorem 4.16.
Let be a relative Rota-Baxter operator on a Leibniz algebra with respect to an antisymmetric representation . Then we have a long exact sequence of cohomology groups:
(56) |
4.4. Some computations of cohomologies
Example 4.17.
Let be the Lie algebra of all matrices over . Then is a relative Rota-Baxter operator on the Lie algebra with respect to the representation , where is defined by
Then is the original Lie algebra . Moreover, is a representation of the Lie algebra , where is defined by
(57) |
Thus, the cohomology of the relative Rota-Baxter operator is the Chevalley-Eilenberg cohomology of with coefficients in the representation . In fact, the Lie algebra is the direct sum , where the factor is the center and consists of diagonal matrices. The Hochschild-Serre formula [22] reads then
The invariant space is zero, because . Furthermore, we have , because all elements are coboundaries. Therefore all the cohomology groups are zero.
Example 4.18.
Let be the Lie algebras of all upper triangular matrices over . Then is a relative Rota-Baxter operator on the Lie algebra with respect to the representation , where is defined by
Thus, the cohomology of the relative Rota-Baxter operator is the Chevalley-Eilenberg cohomology of with coefficients in the representation . Recall that the Lie algebra is an extension
where is the Lie algebras of all upper triangular nilpotent matrices and is the Lie algebras of all diagonal matrices.
The Hochschild-Serre spectral sequence [22] associated to this extension converges towards and has as its second page
Let us compute this -term for . The Lie algebra is -dimensional and we have only non zero contributions from . We have of dimension generated by
From there, we obtain using Dixmier’s Theorem 1 of [17] for the nilpotent Lie algebra with values in a module which does not contains the trivial module
for all . On the other hand, as all maps in are cocycles and the coboundaries correspond to . From there, we obtain again with Dixmier’s Theorem 1
for all . Thus again all the cohomology groups are zero. For general , the whole cohomology will still vanish in case the cohomology spaces do not contain the trivial -module.
Example 4.19.
Let be the Lie algebras of all upper triangular nilpotent matrices over . Then is a relative Rota-Baxter operator on the Lie algebra with respect to the representation , where is defined by
Thus, the cohomology of the relative Rota-Baxter operator is the Chevalley-Eilenberg cohomology of with coefficients in the representation . Since is a nilpotent Lie algebra and the -module does contain the trivial -module (namely the submodule generated by the matrix with non-zero entry in the upper right corner), we have by Dixmier’s Theorem 2 of [17] that
and
Example 4.20.
([41]) Let be a Leibniz algebra. Then the natural projection
gives a relative averaging operator on the Lie algebra with respect to the representation , where is defined by
Then is the original Leibniz algebra . Moreover, is a representation of the Leibniz algebra , where is defined by
Thus, the cohomology of the relative averaging operator is the Loday-Pirashvili cohomology of with coefficients in the antisymmetric representation .
Let now be a finite-dimensional semisimple Leibniz algebra over a field of characteristic zero. By Theorem 4.3 of [18], we obtain that for and an exact sequence
As is already antisymmetric and , the first (non-trivial) map in this sequence is an isomorphism. Furthermore , because is semisimple. Therefore the last (non-trivial) map in this sequence is also an isomorphism (as in Lemma 1.4 of [18]).
Acknowledgements. This research is supported by NSFC (11922110,12001228).
References
- [1] M. Aguiar, Pre-Poisson algebras. Lett. Math. Phys. 54 (2000), 263-277.
- [2] M. Ammar and N. Poncin, Coalgebraic approach to the Loday infinity category, stem differential for -ary graded and homotopy algebras. Ann. Inst. Fourier (Grenoble). 60 (2010), 355-387.
- [3] C. Bai, O. Bellier, L. Guo and X. Ni, Spliting of operations, Manin products and Rota-Baxter operators. Int. Math. Res. Not. (2013), 485-524.
- [4] D. Balavoine, Deformations of algebras over a quadratic operad. Contemp. Math. AMS, 202 (1997), 207-234.
- [5] C. Bai, A unified algebraic approach to the classical Yang-Baxter equation. J. Phys. A: Math. Theor. 40 (2007), 11073-11082.
- [6] C. Bai, L. Guo and X. Ni, Nonabelian generalized Lax pairs, the classical Yang-Baxter equation and PostLie algebras. Comm. Math. Phys. 297 (2010), 553-596.
- [7] C. Barnett, Averaging operators in noncommutative spaces. I. Glasgow Math. J. 24 (1983), 71-74.
- [8] G. Baxter, An analytic problem whose solution follows from a simple algebraic identity. Pacific J. Math. 10 (1960), 731-742.
- [9] A. Bloh, On a generalization of the concept of Lie algebra. (Russian) Dokl. Akad. Nauk SSSR 165 (1965), 471-473.
- [10] R. Bonezzi and O. Hohm, Leibniz gauge theories and infinity structures. Comm. Math. Phys. 377 (2020), 2027-2077.
- [11] M. Bordemann and F. Wagemann, Global integration of Leibniz algebras. J. Lie Theory 27 (2017), 555-567.
- [12] B. Brainerd, On the structure of averaging operators. J. Math. Anal. Appl. 5 (1962), 347-377.
- [13] A. Connes and D. Kreimer, Renormalization in quantum field theory and the Riemann-Hilbert problem. I. The Hopf algebra structure of graphs and the main theorem. Comm. Math. Phys. 210 (2000), 249-273.
- [14] S. Covez, The local integration of Leibniz algebras. Ann. Inst. Fourier (Grenoble) 63 (2013), 1-35.
- [15] A. Das, Deformations of associative Rota-Baxter operators. J. Algebra 560 (2020), 144-180.
- [16] B. Dherin and F. Wagemann, Deformation quantization of Leibniz algebras. Adv. Math. 270 (2015), 21-48.
- [17] J. Dixmier, Cohomologie des algèbres de Lie nilpotentes. Acta Sci. Math. (Szeged) 16 (1955), 246-25
- [18] J. Feldvoss and F. Wagemann, On Leibniz cohomology. J. Algebra 569 (2021), 276-317.
- [19] A. Fialowski and A. Mandal, Leibniz algebra deformations of a Lie algebra. J. Math. Phys. 49 (2008), 093511, 11 pp.
- [20] M. E. Goncharov and P. S. Kolesnikov, Simple finite-dimensional double algebras. J. Algebra 500 (2018), 425-438.
- [21] L. Guo, An introduction to Rota-Baxter algebra. Surveys of Modern Mathematics, 4. International Press, Somerville, MA; Higher Education Press, Beijing, 2012. xii+226 pp.
- [22] G. Hochschild and J. P. Serre, Cohomology of Lie algebras. Ann. of Math. (2) 57 (1953), 591-603.
- [23] O. Hohm and H. Samtleben, Leibniz-Chern-Simons theory and phases of exceptional field theory. Comm. Math. Phys. 369 (2019), 1055-1089.
- [24] C. B. Huijsmans and B. de Pagter, Averaging operators and positive contractive projections. J. Math. Anal. Appl. 113 (1986), 163-184.
- [25] M. Kinyon and A. Weinstein, Leibniz algebras, Courant algebroids, and multiplications on reductive homogeneous spaces. Amer. J. Math. 123 (2001), 525-550.
- [26] P. S. Kolesnikov, Homogeneous averaging operators on semisimple Lie algebras. Algebra Logic 53 (2014), 510-511.
- [27] Y. Kosmann-Schwarzbach, From Poisson algebras to Gerstenhaber algebras. Ann. Inst. Fourier (Grenoble) 46 (1996), 1243-1274.
- [28] A. Kotov and T. Strobl, The embedding tensor, Leibniz-Loday algebras, and their higher gauge theories. Comm. Math. Phys. 376 (2020), 235-258.
- [29] B. A. Kupershmidt, What a classical -matrix really is. J. Nonlinear Math. Phys. 6 (1999), 448-488.
- [30] A. Lazarev, Y. Sheng and R. Tang, Deformations and homotopy theory of relative Rota-Baxter Lie algebras. Comm. Math. Phys. 383 (2021), 595-631.
- [31] J. L. Loday, Une version non commutative des algèbres de Lie: les algèbres de Leibniz. Enseign. Math. (2) 39 (1993), 269-293.
- [32] J. L. Loday and T. Pirashvili, The tensor category of linear maps and Leibniz algebras. (English summary) Georgian Math. J. 5 (1998), 263-276.
- [33] J. L. Loday and T. Pirashvili, Universal enveloping algebras of Leibniz algebras and (co)homology. Math. Ann. 296 (1993), 139-158.
- [34] J. L. Loday, Algebraic K-theory and the conjectural Leibniz K-theory. K-Theory 30 (2003), 105-127.
- [35] J. Pei, C. Bai and L. Guo, Splitting of Operads and Rota-Baxter Operators on Operads. Appl. Categor. Struct. 25 (2017), 505-538.
- [36] J. Pei, C. Bai, L. Guo and X. Ni, Replicating of binary operads, Koszul duality, Manin products and average operators. New Trends in Algebras and Combinatorics (2020), 317-353.
- [37] J. Pei and L. Guo, Averaging algebras, Schroder numbers, rooted trees and operads. J. Algebraic Combin. 42 (2015), 73-109.
- [38] T. Pirashvili, On Leibniz homology. Ann. Inst. Fourier (Grenoble) 44 (1994), 401-411.
- [39] G. C. Rota, Ten mathematics problems I will never solve. Mitt. Dtsch. Math. Ver. 2 (1998), 45-52.
- [40] M. Semonov-Tian-Shansky, What is a classical -matrix? Funct. Anal. Appl. 17 (1983), 259-272.
- [41] Y. Sheng, R. Tang and C. Zhu, The controlling -algebras, cohomologies and homotopy of embedding tensors and Lie-Leibniz triples. Comm. Math. Phys. 386 (2021), 269-304.
- [42] T. Strobl, Leibniz-Yang-Mills gauge theories and the 2-Higgs mechanism. Phys. Rev. D 99 (2019), 115026.
- [43] T. Strobl and F. Wagemann, Enhanced Leibniz algebras: structure theorem and induced Lie 2-algebra. Comm. Math. Phys. 376 (2020), 51-79.
- [44] R. Tang, C. Bai, L. Guo and Y. Sheng, Deformations and their controlling cohomologies of -operators. Comm. Math. Phys. 368 (2019), 665-700.
- [45] R. Tang, C. Bai, L. Guo and Y. Sheng, Homotopy Rota-Baxter operators and post-Lie algebras. J. Noncommut. Geom. 17 (2023), 1-35.
- [46] R. Tang and Y. Sheng, Leibniz bialgebras, relative Rota-Baxter operators and the classical Leibniz Yang-Baxter equation. J. Noncommut. Geom. 16 (2022), 1179-1211.
- [47] R. Tang, Y. Sheng and Y. Zhou, Deformations of relative Rota-Baxter operators on Leibniz algebras. Int. J. Geom. Methods Mod. Phys. 17 (2020), 2050174, 21 pp.
- [48] K. Uchino, Quantum analogy of Poisson geometry, related dendriform algebras and Rota-Baxter operators. Lett. Math. Phys. 85 (2008), 91-109.
- [49] K. Uchino, Twisting on associative algebras and Rota-Baxter type operators. J. Noncommut. Geom. 4 (2010), 349-379.
- [50] K. Uchino, Derived bracket construction and Manin products. Lett. Math. Phys. 93 (2010), 37-53.
- [51] B. Vallette, Manin products, Koszul duality, Loday algebras and Deligne conjecture. J. Reine Angew. Math. 620 (2008), 105-164.
- [52] K. Wang and G. Zhou, Cohomology theory of averaging algebras, -structures and homotopy averaging algebras. arXiv:2009.11618.