Categorical dualtiy for Yetter–Drinfeld C ∗ -algebras Beyond the braided-commutative case
Lucas Hataishi, Makoto Yamashita · 2025
Read Categorical dualtiy for Yetter–Drinfeld C ∗ -algebras Beyond the braided-commutative case by Lucas Hataishi, Makoto Yamashita (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-2504-21162.txt.
Introduction
Quantum groups twist the symmetries of classical physics, and their algebras carry two compatible actions at once. Lucas Hataishi and Makoto Yamashita study the duality theory for these Yetter-Drinfeld algebras in the C*-algebra setting. Earlier results covered only a special braided-commutative case. This work goes beyond it.
In the spirit of Tannaka-Krein duality, they show that such algebras correspond exactly to certain bimodule categories over the quantum group's representation category. A centrality condition picks out the right ones. The correspondence lets each side be rebuilt from the other. It gives operator algebraists a categorical dictionary for quantum symmetry.
Catalog record
| CREATOR | Lucas Hataishi, Makoto Yamashita |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 83 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-21162
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Categorical dualtiy for Yetter–Drinfeld C ∗ -algebras Beyond the braided-commutative case Thanks: Supported by the NFR project 300837 “Quantum Symmetry”. Lucas Hataishi, Makoto Yamashita Abstract . We develop a tensor categorical duality in the sprit of the Tannaka–Krein duality for the C ∗ -algebras admitting the Yetter–Drinfeld module structure over a compact quantum group. Under this duality, given a reduced compact quantum group G G , the Yetter–Drinfeld G G -C ∗ -algebras correspond to the bimodule categories over the representation category Rep ( G ) \Rep(G) , satisfying a certain centrality condition. \DefineSimpleKey bibhow 1. Introduction The Tannaka–Krein duality principle for quantum groups has been an effective tool to compare different incarnations of quantum symmetries . In the original form by Woronowicz [ 16 ] , it states that compact quantum groups are essentially the same as a pair of a rigid C ∗ -tensor category and a tensor functor into the category of finite dimensional Hilbert spaces. More recently, this was generalized to actions of quantum groups, which are modeled by comodule algebras on the algebraic side, and module categories on the tensor categorical side. In the operator algebraic setting, Tannaka–Krein theory is mostly developed for compact quantum groups G , as the complete reducibility of the representations of G allows us to control possibly infinite-dimensional algebras with an action of G in an algebraic way. While a straightforward generalization of this to locally compact quantum groups seems hopeless, there is one promising class of locally compact quantum groups with compact-by-discrete type structure like totally disconnected groups: the Drinfeld double quantum group DG , which can be considered as a matched pair of a compact quantum group G and its discrete dual quantum group \hat{G} . Actions of DG on C ∗ -algebras can be described as actions of G with additional structure encoding the action of \hat{G} . Since G -actions can be described categorically, one might hope to describe DG -actions categorically as well, and obtain a Tannaka–Krein duality for DG -actions. This program was initiated in [ 8 ] , where Neshveyev and the second named author found that embeddings of the C ∗ -tensor category \Rep(G) of finite dimensional unitary representations of G into other C ∗ -tensor categories correspond to continuous unital DG -C ∗ -algebras having braided-commutativity . In this paper, we remove this braided commutativity assumption, obtaining a categorical description of general continuous unital Yetter–Drinfeld C ∗ -algebras. Specifically, we establish a categorical equivalence between the category of such algebras and the category of certain bimodule categories over the representation category \Rep(G) . A similar duality statement was recently given in [ 4 ] at the level weak monoidal functors developing the line of works by Pinzari–Roberts [ 15 ] and Neshveyev [ 12 ] . In a follow-up paper [ 6 ] , we will use the results of this paper to show that the theory of boundary actions for Drinfeld doubles of compact quantum groups is invariant under monoidal equivalence. In particular, for two monoidally equivalent compact quantum groups G and G^{\prime} , the Furstenberg-Hamana boundary \partial_{FH}(DG) of DG is trivial if and only if \partial_{FH}(DG^{\prime}) is trivial. This answers a question left open in [ 5 ] . Another potential application of the results in this paper is in geometric representation theory for quantum groups. There is a non-trivial relation between the topology of surfaces and module categories over braided tensor categories [ 2 ] . The category of unitary representations of the Drinfeld double DG of a compact quantum group G is the unitary Drinfeld center \mathcal{Z}(\mathrm{Hilb}(\Rep(G))) , where \mathrm{Hilb}(\Rep(G)) is the category of all unitary representations of G , i.e., it allows for infinite dimensional representations, see [ 9 ] . Now let us describe the content of the paper. In Section 2 , we collect the preliminary materials on rigid C ∗ -tensor categories, compact quantum groups, and their Drinfeld doubles. In Section 3 , we introduce what will later be shown to be the Tannaka–Krein categorical dual of continuous unital Yetter–Drinfeld C ∗ -algebras. In order to account for the extra structures and properties of Yetter–Drinfeld C ∗ -algebras, we consider bimodule categories (\mathcal{M},m) over \Rep(G) with a distinguished \Rep(G) -generator m , equipped with unitary ismorphisms \sigma_{U}\colon U\stmryogt m\to m\stmryolt U,\quad(U\in\Rep(G)), which are natural in U and satisfy a half-braided condition: \sigma_{U\otimes V}=(\sigma_{U}\stmryolt\id_{V})\circ(\id_{U}\stmryogt\sigma_{V}). We call such categories centrally pointed cyclic bimodule C ∗ - \Rep(G) -module categories. We prove our main result in Section 4 , giving the equivalence between the category of continuous unital Yetter–Drinfeld C ∗ -algebras and the category of centrally pointed cyclic C ∗ - \Rep(G) -bimodule categories. It is based on the following observation. Suppose B is a Yetter–Drinfeld C ∗ -algebra. The extra categorical structure we are looking for boils down to having a G -equivariant B -bimodule structure on the G -equivariant right B -module H_{U}\otimes B , where H_{U} is the underlying finite-dimensional Hilbert space for an object U in \Rep(G) . Let us write the left action \beta\colon\mathcal{O}(G)\odot B\to B of the regular subalgebra \mathcal{O}(G)\subset C(G) as \beta(x\odot b)=x\triangleright b , and write U as U=\sum_{ij}m^{U}_{ij}\otimes u_{ij}\in\mathcal{B}(H_{U})\otimes C(G). We can define a left B -action \pi_{U}\colon B\to\End_{B}(H_{U}\otimes B) on H_{U}\otimes B by \pi_{U}(b)=\sum_{ij}m^{U}_{ij}\otimes(u_{ij}\triangleright b). In [ 8 ] , the braided-commutativity assumption was used to give an alternative description of this action, allowing the proof that it indeed defines a G -equivariant * -homomorphism. In Proposition 4.3 , we show that braided-commutativity is not needed, proving that \pi_{U} is G -equivariant using only the relation between the coaction of \mathcal{O}(G) and \beta fitting in a Yetter–Drinfeld module. 2. Preliminaries 2.1. C ∗ -tensor categories Here we mostly follow [ 7 ] [ 8 ] . Given a C ∗ -category \mathcal{C} , we denote the space of morphisms from an object X to another Y as \mathcal{C}(X,Y) . The involution \mathcal{C}(X,Y)\to\mathcal{C}(Y,X) is denoted by T\mapsto T^{*} , and the norm is by \left\|T\right\| , so that we have the C ∗ -identity \left\|T^{*}T\right\|=\left\|T\right\|^{2} . We also assume that \mathcal{C} has finite direct sums. Given X,Y\in\mathcal{C} , their direct sum is denoted by X\oplus Y and there are isometries \iota_{X}\colon X\to X\oplus Y\leftarrow Y\penalty\mskip 6.0mu plus 1.0mu\mathpunct{}\nonscript\mkern-3.0mu{:}\mskip 2.0mu\iota_{Y} such that \iota_{X}\circ\iota_{X}^{*}+\iota_{Y}\circ\iota_{Y}^{*}=\id_{X\oplus Y} . Under these assumptions, \mathcal{C}(X,Y) is naturally a right Hilbert module over the unital C ∗ -algebra \mathcal{C}(X)=\mathcal{C}(X,X) . We tacitly assume that \mathcal{C} is closed under taking subobjects, i.e., any projection in \mathcal{C}(X) corresponds to a direct summand of X . A C ∗ -tensor category is a C ∗ -category endowed with monoidal structure given by a * -bifunctor \otimes\colon\mathcal{C}\times\mathcal{C}\to\mathcal{C} , a unit object 1_{\mathcal{C}} , and unitary natural isomorphisms 1_{\mathcal{C}}\otimes U\to U\leftarrow U\otimes 1_{\mathcal{C}},\qquad\Phi\colon(U\otimes V)\otimes W\to U\otimes(V\otimes W) for U,V,W\in\mathcal{C} , satisfying a standard set of axioms. Without losing generality, we may and do assume that \mathcal{C} is strict so that the above morphisms are identity, and that 1_{\mathcal{C}} is simple, unless explicitly stated otherwise. A rigid C ∗ -tensor category is a C ∗ -tensor category where any object U has a dual given by: an object \bar{U} and morphisms R\colon 1_{\mathcal{C}}\to\bar{U}\otimes U,\qquad\bar{R}\colon 1_{\mathcal{C}}\to U\otimes\bar{U} satisfying the conjugate