Searcharxiv⌕ Search

arXiv subjects

R. Exel

Publications and source records attributed to R. Exel.

At least 19 recordsLinked to original sources
arXiv2025

Twisted Steinberg algebras, regular inclusions and induction

Given a field $K$ and an ample (not necessarily Hausdorff) groupoid $G$, we define the concept of a line bundle over $G$ inspired by the well known concept from the theory of C*-algebras. If $E$ is such a line bundle, we construct the associated twisted Steinberg algebra in terms of sections of $E$, extending the original construction introduced independently by Steinberg in 2010, and by Clark, Farthing, Sims and Tomforde in a 2014 paper (originally announced in 2011). We also generalize (strictly, in the non-Hausdorff case) the 2023 construction of (cocycle) twisted Steinberg algebras of Armstrong, Clark, Courtney, Lin, Mccormick and Ramagge. We then extend Steinberg's theory of induction of modules, not only to the twisted case, but to the much more general case of regular inclusions of algebras. Among our main results, we show that, under appropriate conditions, every irreducible module is induced by an irreducible module over a certain abstractly defined isotropy algebra. We also describe a process of disintegration of modules and use it to prove a version of the Effros-Hahn conjecture, showing that every primitive ideal coincides with the annihilator of a module induced from isotropy.

math.OA↗
arXiv2022

Strong equivalence of graded algebras

We introduce the notion of a strong equivalence between graded algebras and prove that any partially-strongly-graded algebra by a group $G$ is strongly-graded-equivalent to the skew group algebra by a product partial action of $G$. As to a more general idempotent graded algebra $B$, we point out that the Cohen-Montgomery duality holds for $B$, and $B$ is graded-equivalent to a global skew group algebra. We show that strongly-graded-equivalence preserves strong gradings and is nicely related to Morita equivalence of product partial actions. Furthermore, we prove that any product partial group action $\alpha $ is globalizable up to Morita equivalence; if such a globalization $\beta $ is minimal, then the skew group algebras by $\alpha $ and $\beta $ are graded-equivalent; moreover, $\beta $ is unique up to Morita equivalence. Finally, we show that strongly-graded-equivalent partially-strongly-graded algebras are stably isomorphic as graded algebras.

math.RA↗
arXiv2019

Subshift semigroups

Given a one-sided subshift $X$ on a finite alphabet, we consider the semigroup $S_X =L_X \cup \{0\}$, where $L_X $ is the language of $X $, equipped with the multiplication operation given by concatenation, when allowed, and set to vanish otherwise. We then study the inverse hull $H(S_X )$, relating it with C*-algebras that have been discussed in the literature in association with subshifts.

math.OA↗
arXiv2019

Characterizing groupoid C*-algebras of non-Hausdorff \'etale groupoids

Given a not-necessarily Hausdorff, topologically free, twisted \'etale groupoid $(G, L)$, we consider its "essential groupoid C*-algebra", denoted $C^*_{ess}(G, L)$, obtained by completing $C_c(G, L)$ with the smallest among all C*-seminorms coinciding with the uniform norm on $C_c(G^0)$. The inclusion of C*-algebras $(C_0(G^0), C^*_{ess}(G, L))$ is then proven to satisfy a list of properties characterizing it as what we call a "weak Cartan inclusion". We then prove that every weak Cartan inclusion $(A, B)$, with $B$ separable, is modeled by a topologically free, twisted \'etale groupoid, as above. In our second main result we give a necessary and sufficient condition for an inclusion of C*-algebras $(A, B)$ to be modeled by a twisted \'etale groupoid based on the notion of "canonical states". A simplicity criterion for $C^*_{ess}(G, L)$ is proven and many examples are provided.

math.OA↗
arXiv2018

Higher rank graphs, k-subshifts and k-automata

Given a $k$-graph $\Lambda $ we construct a Markov space $M_\Lambda $, and a collection of $k$ pairwise commuting cellular automata on $M_\Lambda $, providing for a factorization of Markov's shift. Iterating these maps we obtain an action of ${\mathbb N}^k$ on $M_\Lambda $ which is then used to form a semidirect product groupoid $M_\Lambda \rtimes {\mathbb N}^k$. This groupoid turns out to be identical to the path groupoid constructed by Kumjian and Pask, and hence its C*-algebra is isomorphic to the higher rank graph C*-algebra of $\Lambda $.

math.OA↗
arXiv2018

Representations of the inverse hull of a 0-left cancellative semigroup

A semigroup S containing a zero element is said to be 0-left cancellative if st = sr \neq 0 implies that t = r. Given such an S we build an inverse semigroup H(S), called the inverse hull of S. Motivated by the study of certain C*-algebras associated to H(S) (a task that we will address in a subsequent article) we carry out a detailed analysis of the spectrum of the idempotent semilattice E(S) of H(S) with a special interest in identifying the ultra-characters. In order to produce examples of characters on E(S), we introduce the notion of "strings" in a semigroup, attempting to make sense of the "infinite paths" which are fundamental in the study of graph C*-algebras. Our strongest results are obtained under the assumption that S admits "least common multiples", but we also touch upon the notion of "finite alignment", motivated by the corresponding notion from the theory of higher rank graphs, and which has also appeared in recent papers by Spielberg and collaborators.

math.OA↗
arXiv2017

The inverse hull of 0-left cancellative semigroups

Given a semigroup S with zero, which is left-cancellative in the sense that st=sr \neq 0 implies that t=r, we construct an inverse semigroup called the inverse hull of S, denoted H(S). When S admits least common multiples, in a precise sense defined below, we study the idempotent semilattice of H(S), with a focus on its spectrum. When S arises as the language semigroup for a subsift X on a finite alphabet, we discuss the relationship between H(S) and several C*-algebras associated to X appearing in the literature.

math.OA↗
arXiv2016

The ideal structure of algebraic partial crossed products

Given a partial action of a discrete group $G$ on a Hausdorff, locally compact, totally disconnected topological space $X$, we consider the correponding partial action of $G$ on the algebra $L_c(X)$ consisting of all locally constant, compactly supported functions on $X$, taking values in a given field $K$. We then study the ideal structure of the algebraic partial crossed product $L_c(X)\rtimes G$. After developping a theory of induced ideals, we show that every ideal in $L_c(X)\rtimes G$ may be obtained as the intersection of ideals induced from isotropy groups, thus proving an algebraic version of the Effros-Hahn conjecture.

math.OA↗
arXiv2015

Partial actions and subshifts

Given a finite alphabet $\Lambda$, and a not necessarily finite type subshift $X\subseteq \Lambda^\infty$, we introduce a partial action of the free group $F(\Lambda)$ on a certain compactification $\Omega_X$ of $X$, which we call the spectral partial action. The space $\Omega_X$ has already appeared in many papers in the subject, arising as the spectrum of a commutative C*-algebra usually denoted by ${\cal D}_X$. Since the descriptions given of $\Omega_X$ in the literature are often somewhat terse and obscure, one of our main goals is to present a sensible model for it which allows for a detailed study of its structure, as well as of the spectral partial action, from various points of view, including topological freeness and minimality. We then apply our results to study certain C*-algebras associated to $X$, introduced by Matsumoto and Carlsen. Most of the results we prove are already well known, but our proofs are hoped to be more natural and more in line with mainstream techniques used to treat similar C*-algebras. The clearer understanding of $\Omega_X$ provided by our model in turn allows for a fine tuning of some of these results, including a necessary and sufficient condition for the minimality of the Carlsen-Matsumoto C*-algebra ${\cal O}_X$, generalizing a similar result of Thomsen.

math.OA↗
arXiv2012

Blends and Alloys

Given two algebras A and B, sometimes assumed to be C*-algebras, we consider the question of putting algebra or C*-algebra structures on the tensor product A\otimes B. In the C*-case, assuming B to be two-dimensonal, we characterize all possible such C*-algebra structures in terms of an action of the cyclic group Z_2. An example related to commuting squares is also discussed.

math.OA↗
arXiv2010

Globalization of twisted partial actions

Let A be a unital ring which is a product of possibly infinitely many indecomposable rings. We establish criteria for the existence of a globalization for a given twisted partial action of a group on A. If the globalization exists, it is unique up to a certain equivalence relation and, moreover, the crossed product corresponding to the twisted partial action is Morita equivalent to that corresponding to its globalization. For arbitrary unital rings the globalization problem is reduced to an extendibility property of the multipliers involved in the twisted partial action.

math.RA↗
arXiv2008

Crossed products by twisted partial actions and graded algebras

For a twisted partial action Θof a group G on an (associative non-necessarily unital) algebra A over a commutative unital ring k, the crossed product A X_ΘG is proved to be associative. Given a G-graded k-algebra B = \oplus_{g\in G}\B_g with the mild restriction of homogeneous non-degeneracy, a criteria is established for B to be isomorphic to the crossed product B_1 X_ΘG for some twisted partial action of G on B_1. The equality B_g\B_{g^{-1}}B_g = \B_g for all g\in G is one of the ingredients of the criteria, and if it holds and, moreover, B has enough local units, then it is shown that B is stably isomorphic to a crossed product by a twisted partial action of G.

math.RA↗
arXiv2008

Envelope Algebras of Partial Actions as Groupoid C*-Algebras

We describe the envelope C*-algebra associated to a partial action of a countable discrete group on a locally compact space as a groupoid C*-algebra (more precisely as a C*-algebra from an equivalence relation) and we use our approach to show that, for a large class of partial actions of Z on the Cantor set, the envelope C*-algebra is an AF-algebra. We also completely characterize partial actions of a countable discrete group on a compact space such that the envelope action acts in a Hausdorff space.

math.OA↗
arXiv2006

Associativity of crossed products by partial actions, enveloping actions and partial representations

Given a partial action αof a group G on an associative algebra A we consider the crossed product A x_αG. Using the algebras of multipliers of ideals of A we prove that A x_αG is associative, provided that all ideals of A are idempotent. This generalizes a previous result on the associativity of A x_αG in the context of C*-algebras. We also give a criteria for the existence of a global extension of a given partial action on an algebra and use crossed products to study relations between partial actions of groups on algebras and partial representations. As an application we endow partial group algebras with crossed product structure.

math.RA↗
arXiv2004

The Crossed Product by a Partial Endomorphism

Given a closed ideal I in a C*-algebra A, an ideal J (not necessarily closed) in I, a *-homomorphism \al:A --> M(I) and a map L:J --> A with some properties, based on [3] and [9] we define a C*-algebra O(A,\al,L) which we call the "Crossed Product by a Partial Endomorphism." In the second section we introduce the Crossed Product by a Partial Endomorphism O(X,\al,L) induced by a local homeomorphism σ:U --> X where X is a compact Hausdorff space and U is an open subset of X.The main result of this section is that every nonzero gauge invariant ideal of O(X,\al,L) has nonzero intersection with C(X). We present the example which motivated this work, the Cuntz-Krieger algebra for infinite matrices. We show in the third section a bijection between the gauge invariant ideals of O(X,\al,L) and the σ,σ^{-1} -invariant open subsets of X. The last section is dedicated to the study of O(X,\al,L) in the case where the pair (X,σ) has an extra property, wich we call topological freeness. We prove that in this case every nonzero ideal of O(X,\al,L) has nonzero intersection with C(X). If moreover (X,σ) has the property that (X',σ_X') is topologically free for each closed sigma,σ^{-1}-invariant subset X' of X then we obtain a bijection between the ideals of O(X,\al,L) and the open σ,σ^{-1}-invariant subsets of X. We conclude this section by showing a simplicity criteria for the Cuntz-Krieger algebras for infinite matrices.

math.OA↗
arXiv2003

AF-algebras and the tail-equivalence relation on Bratteli diagrams

Given a Bratteli diagram D we consider the compact topological space formed by all infinite paths on D. Two such path are said to be tail-equivalent when they "have the same tail", i.e. when they eventually coincide. This equivalence relation is approximately proper and hence one may consider the C*-algebra associated to it according to a procedure recently introduced by the first named author and A. Lopes. The main result of this work is the proof that this algebra is isomorphic to the AF-algebra associated to the given Bratteli diagram.

math.OA↗