Radial restriction of spherical functions on supergroups

Mitra Mansouri and Hadi Salmasian · 2025

Read Radial restriction of spherical functions on supergroups by Mitra Mansouri and Hadi Salmasian (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-19102.txt.

Introduction

Mitra Mansouri and Hadi Salmasian give a pure algebra treatment of a class of functions on groups that mix signs, known as supergroups. Their paper is from two thousand twenty-five.

The functions they study stay the same when shifted on both sides by a sub-group. The team builds them as a part of a dual algebra, cut out by a set ideal. For a broad class of paired super structures, they define how to trim such a function down to a core subspace. They prove that trim loses no facts. For one set pair, they compute a base for the ideal and find a link to two famed count lines.

Catalog record

CREATORMitra Mansouri and Hadi Salmasian
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE57 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-19102

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Radial restriction of spherical functions on supergroups Mitra Mansouri and Hadi Salmasian Abstract . Using the Hopf superalgebra structure of the enveloping algebra U ⁡ ( 𝔤 ) U(\mathfrak{g}) of a Lie superalgebra 𝔤 = Lie ⁡ ( G ) \mathfrak{g}=\mathrm{Lie}(G) , we give a purely algebraic treatment of K K -bi-invariant functions on a Lie supergroup G G , where K K is a sub-supergroup of G G . We realize K K -bi-invariant functions as a subalgebra 𝒜 ⁡ ( 𝔤 , 𝔨 ) \mathcal{A}(\mathfrak{g},\mathfrak{k}) of the dual of U ⁡ ( 𝔤 ) U(\mathfrak{g}) whose elements vanish on the coideal ℐ = 𝔨 ​ U ​ ( 𝔤 ) + U ⁡ ( 𝔤 ) ​ 𝔨 \mathcal{I}=\mathfrak{k}U(\mathfrak{g})+U(\mathfrak{g})\mathfrak{k} , where 𝔨 = Lie ⁡ ( K ) \mathfrak{k}=\mathrm{Lie}(K) . Next, for a general class of supersymmetric pairs ( 𝔤 , 𝔨 ) (\mathfrak{g},\mathfrak{k}) , we define the radial restriction of elements of 𝒜 ⁡ ( 𝔤 , 𝔨 ) \mathcal{A}(\mathfrak{g},\mathfrak{k}) and prove that it is an injection into S ​ ( 𝔞 ) ∗ S(\mathfrak{a})^{*} , where 𝔞 \mathfrak{a} is the Cartan subspace of ( 𝔤 , 𝔨 ) (\mathfrak{g},\mathfrak{k}) . Finally, we compute a basis for ℐ \mathcal{I} in the case of the pair ( 𝔤 ​ 𝔩 ​ ( 1 | 2 ) , 𝔬 ​ 𝔰 ​ 𝔭 ​ ( 1 | 2 ) ) (\mathfrak{gl}(1|2),\mathfrak{osp}(1|2)) , and uncover a connection with the Bernoulli and Euler zigzag numbers. 1. Introduction A spherical function on a real reductive Lie group G is a K -bi-invariant eigenfunction of the algebra of invariant differential operators on G/K , where K is a maximal compact subgroup of G . The theory of spherical functions is extensive and has profound connections with representation theory and combinatorics of symmetric functions [ 5 , 7 , 3 ] . However, in the context of supergroups this theory is far less developed. This is partly due to the cumbersome technicalities that arise in general in superanalysis and supergeometry, and create major hurdles in extending the tools of harmonic analysis to the super setting. To circumvent the above technical difficulty, we pursue an indirect, purely algebraic approach to study spherical functions on supersymmetric spaces. This approach was first explored by Sergeev [ 12 ] for two families of symmetric pairs (\mathfrak{gl}(m|n)\oplus\mathfrak{gl}(m|n),\mathfrak{gl}(m|n)) and (\mathfrak{gl}(m|2n),\mathfrak{osp}(m|2n)) . An extended, yet unpublished, version of [ 12 ] is available in [ 11 ] . However, Sergeev addressed each of the above families by a different and explicit method. From this viewpoint, the papers [ 12 , 11 ] use ad hoc methods. In this paper, we expand on the idea of Sergeev in the general setting of supersymmetric pairs (\mathfrak{g},\mathfrak{k}) of Lie superalgebras. Let \mathcal{A}(\mathfrak{g},\mathfrak{k}) denote the dual of the quotient U(\mathfrak{g})/\mathcal{I} where \mathcal{I}:=\mathfrak{k}U(\mathfrak{g})+U(\mathfrak{g})\mathfrak{k}. Since \mathcal{I} is a coideal, \mathcal{A}(\mathfrak{g},\mathfrak{k}) is a (super)commutative algebra. Elements of \mathcal{A}(\mathfrak{g},\mathfrak{k}) play the role of K -bi-invariant functions on G . Of course the Lie algebra \mathfrak{g} does not contain the information about isogeny of G . But in some sense \mathcal{A}(\mathfrak{g},\mathfrak{k}) contains all analytic spherical functions on all isogenous pairs (G,K) associated to (\mathfrak{g},\mathfrak{k}) , in the same way that U(\mathfrak{g})^{*} contains all analytic functions on the isogeny class of G in the non-super setting (where we identify functions by their derivatives at identity). Our main result is Theorem 3.9 , which states that K -bi-invariant functions are uniquely determined by their radial restriction. This is a variant of a well-known fact from the purely even case (see Remark 3.10 ), and leads to interesting problems that we aim to explore in the future. Theorem 3.9 was also proved in [ 10 , Theorem 4.1] for several families of pairs (\mathfrak{g},\mathfrak{k}) , under the assumption that \mathfrak{g} has a non-degenerate invariant form that restricts to a non-degenerate form on the Cartan subspace \mathfrak{a} of \mathfrak{g} . Our proof has a small advantage over the one given in [ 10 ] : it does not use the invariant form of \mathfrak{g} , and therefore it applies to a larger class of pairs (\mathfrak{g},\mathfrak{k}) ; see Example 3.11 . We conclude this paper by looking more closely at the supersymmetric pair (\mathfrak{gl}(1|2),\mathfrak{osp}(1|2)) . We compute a basis of \mathcal{I}=\mathfrak{k}U(\mathfrak{g})+U(\mathfrak{g})\mathfrak{k} in Theorem 4.8 . The interesting observation is that Bernoulli numbers occur as coefficients in our basis. Along the way, we also come across the Euler zigzag numbers. Acknowledgement. We thank the anonymous referee for drawing our attention to the work of Sergeev and Veselov in [ 10 ] . This paper is based on the doctoral dissertation of the first author, which was completed under supervision of the second author. Certain aspects of this project were refined through discussions with the participants of the AIM SQuaREs meetings on Symmetric spaces and Capelli operators for Lie superalgebras. The research of the second author is partially supported by NSERC Discovery Grant RGPIN-2024-04030. 2. The algebra \mathcal{A}(\mathfrak{g},\mathfrak{k}) Throughout this paper, the base field will be \mathbb{C} . Let \mathfrak{g} be any Lie superalgebra. Then the universal enveloping algebra U(\mathfrak{g}) is a Hopf superalgebra, with co-product \Delta:U(\mathfrak{g})\to U(\mathfrak{g})\otimes U(\mathfrak{g}) defined by \Delta(x):=1\otimes x+x\otimes 1 for x\in\mathfrak{g} , antipode defined by S(x)=-x for x\in\mathfrak{g} , and counit defined by \varepsilon(1)=1 and \varepsilon(x)=0 for x\in\mathfrak{g} . Using the comultiplication of U(\mathfrak{g}) , we can equip the dual U(\mathfrak{g})^{*} with an associative algebra structure. The multiplication of U(\mathfrak{g})^{*} is given by \phi\psi(x):=\sum(-1)^{|x_{1}||\psi|}\phi(x_{1})\psi(x_{2})\text{ for $x\in U(\mathfrak{g})$,} where \Delta(x)=\sum x_{1}\otimes x_{2} and as usual |\cdot| denotes parity. Remark 2.1 . Let H be a Hopf (super)algebra. An element of H^{*} is said to be of finite type if its kernel contains a two-sided ideal of H of finite codimension. The elements of finite type of H^{*} form a Hopf algebra (see [ 13 , Ch. VI] ), which we will denote by H^{\circ} . Let (\pi,V) be a finite dimensional \mathfrak{g} -module. Recall that the matrix coefficients of (\pi,V) are linear maps \phi_{v,v^{*}}\in U(\mathfrak{g})^{*} defined by (1) \phi_{v,v^{*}}(x):=\langle v^{*},\pi(x)v\rangle\text{ for all }x\in U(\mathfrak{g}),\ v\in V,\ v^{*}\in V^{*}. If \mathrm{Ann}(V) denotes the annihilator of a \mathfrak{g} -module module V , then U(\mathfrak{g})/\mathrm{Ann}(V) embeds in \mathrm{End}_{\mathbb{C}}(V) , hence \phi_{v,v^{*}}\in U(\mathfrak{g})^{\circ} . Indeed the converse also holds and is straightforward to prove: elements of U(\mathfrak{g})^{\circ} are matrix coefficients of finite dimensional \mathfrak{g} -modules. Let \mathfrak{k}\subseteq\mathfrak{g} be any subalgebra of \mathfrak{g} and set \mathcal{I}:=\mathfrak{k}U(\mathfrak{g})+U(\mathfrak{g})\mathfrak{k} . We say \lambda\in U(\mathfrak{g})^{*} is \mathfrak{k} -bi-invariant if \mathcal{I}\subseteq\ker(\lambda) . Since \mathfrak{k} is \mathbb{Z}_{2} -graded, the \mathfrak{k} -bi-invariant \lambda\in U(\mathfrak{g})^{*} form a graded subspace which we denote by \mathcal{A}(\mathfrak{g},\mathfrak{k}) . Lemma 2.2 . The superspace \mathcal{A}(\mathfrak{g},\mathfrak{k}) is a subalgebra of U(\mathfrak{g})^{*} . Proof. This is immediate from the fact that \mathcal{I} is a coideal of U(\mathfrak{g}) , but we provide more details. Given \lambda,\mu\in\mathcal{A}(\mathfrak{g},\mathfrak{k}) , we want to prove that \lambda\mu\in\mathcal{A}(\mathfrak{g},\mathfrak{k}) . For k\in\mathfrak{k} and u\in U(\mathfrak{g}) , if \Delta(u)=\sum u_{1}\otimes u_{2} in Sweedler’s notation then \displaystyle\lambda\mu(ku)=\lambda\otimes\mu(( \displaystyle k\otimes 1+1\otimes k)\Delta(u))=\lambda\otimes\mu\left(\left(k\otimes 1+1\otimes k\right)\sum u_{1}\otimes u_{2}\right) \displaystyle=\sum\left((-1)^{|\mu|\cdot(|k|+|u_{1}|)}\lambda\left(ku_{1}\right)\otimes\mu\left(u_{2}\right)+(-1)^{(|k|+|\mu|)|u_{1}|}\lambda\left(u_{1}\right)\otimes\mu\left(ku_{2}\right)\right)=0, hence \mathfrak{k}U(\mathfrak{g})\subseteq\ker(\lambda\mu) . Similarly, U(\mathfrak{g})\mathfrak{k}\subseteq\ker(\lambda\mu) , so that \lambda\mu\in\mathcal{A}(\mathfrak{g},\mathfrak{k}) . ∎ The superalgebra U(\mathfrak{g}) acts by left and right translation on itself as follows: (2) L_{x}:y\mapsto xy\quad\text{and}\quad R_{x}:y\mapsto(-1)^{|x|.|y|}yx. Now set \mathcal{A}(\mathfrak{g},\mathfrak{k})^{\circ}:=\mathcal{A}(\mathfrak{g},\mathfrak{k})\cap U(\mathfrak{g})^{\circ}. The justification that \mathcal{A}(\mathfrak{g},\mathfrak{k})^{\circ} plays the role of spherical functions for (\mathfrak{g},\mathfrak{k})

Read the complete text