The formal theory of Hopf algebras Part II: The case of Hopf algebras

Abstract

The category Hopf_R of Hopf algebras over a commutative unital ring R is analyzed with respect to its categorical properties. The main results are: (1) For every ring R the category Hopf_R is locally presentable, it is coreflective in the category of bialgebras over R, over every R-algebra there exists a cofree Hopf algebra. (2) If, in addition, R is absoluty flat, then Hopf_R is reflective in the category of bialgebras as well, and there exists a free Hopf algebra over every R-coalgebra. Similar results are obtained for relevant subcategories of Hopf_R. Moreover it is shown that, for every commutative unital ring R, the so-called “dual algebra functor” has a left adjoint and that, more generally, universal measuring coalgebras exist.     

Weak bimonoids in duoidal categories

Weak bimonoids in duoidal categories are introduced. They provide a common generalization of bimonoids in duoidal categories and of weak bimonoids in braided monoidal categories. Under the assumption that idempotent morphisms in the base category split, they are shown to induce weak bimonads (in four symmetric ways). As a consequence, they have four separable Frobenius base (co)monoids, two in each of the underlying monoidal categories. Hopf modules over weak bimonoids are defined by weakly lifting the induced comonad to the Eilenberg–Moore category of the induced monad. Making appropriate assumptions on the duoidal category in question, the fundamental theorem of Hopf modules is proven which says that the category of modules over one of the base monoids is equivalent to the category of Hopf modules if and only if a Galois-type comonad morphism is an isomorphism.     

A Quillen Model Structure for 2-Categories

We describe a cofibrantly generated Quillen model structure on the locally finitely presentable category 2-Cat of (small) 2-categories and 2-functors; the weak equivalences are the biequivalences, and the homotopy relation on 2-functors is just pseudonatural equivalence. The model structure is proper, and is compatible with the monoidal structure given by the Gray tensor product. It is not compatible with the Cartesian closed structure, in which the tensor product is the product.The model structure restricts to a model structure on the full subcategory PsGpd of 2-Cat, consisting of those 2-categories in which every arrow is an equivalence and every 2-cell is invertible. The model structure on PsGpd is once again proper, and compatible with the monoidal structure given by the Gray tensor product.     

Free Internal Groups

In the first part of this note an elementary proof is given of the fact that algebraic functors, that is, functors induced by morphisms of Lawvere theories, have left adjoints provided that the category K\mathcal{K} in which the models of these theories take their values is locally presentable. The main focus however lies on the special cases of the underlying functor of the category Grp(K)\mathsf{Grp}(\mathcal{K}) of internal groups in K\mathcal{K} and the embedding of Grp(K)\mathsf{Grp}(\mathcal{K}) into Mon(K)\mathsf{Mon}(\mathcal{K}), the category of monoids in K\mathcal{K}: Here a unifying construction of the respective left adjoints is provided which not only works in case K\mathcal{K} is a locally presentable category but also when K\mathcal{K} is, for example, a particular category of topological spaces such as the category of Hausdorff or Tychonoff spaces or a cartesian closed topological category.     

A duality theorem for graph embeddings

A generalized type of graph covering, called a “Wrapped quasicovering” (wqc) is defined. If K, L are graphs dually embedded in an orientable surface S, then we may lift these embeddings to embeddings of dual graphs K?,L? in orientable surfaces S?, such that S? are branched covers of S and the restrictions of the branched coverings to K?,L? are wqc's of K, L. the theory is applied to obtain genus embeddings of composition graphs G[nK₁] from embeddings of “quotient” graphs G.     

Graph expansions of unipotent monoids

Margolis and Meakin use the Cayley graph of a group presentation to construct E-unitary inverse monoids [11]. This is the technique we refer to as graph expansion. In this paper we consider graph expansions of unipotent monoids, where a monoid is unipotent if it contains a unique idempotent. The monoids arising in this way are E-unitary and belong to the quasivariety of weakly left ample monoids. We give a number of examples of such monoids. We show that the least unipotent congruence on a weakly left ample monoid is given by the same formula as that for the least group congruence on an inverse monoid and we investigate the notion of proper for weakly left ample monoids.

Using graph expansions we construct a functor F^e from the category U of unipotent monoids to the category PWLA of proper weakly left ample monoids. The functor F^e is an expansion in the sense of Birget and Rhodes [2]. If we equip proper weakly left ample monoids with an extra unary operation and denote the corresponding category by PWLA ⁰ then regarded as a functor U→PWLA ⁰ F^e is a left adjoint of the functor F^σ : PWLA ⁰ → U that takes a proper weakly left ample monoid to its greatest unipotent image.

Our main result uses the covering theorem of [8] to construct free weakly left ample monoids.     

Schur-Finiteness and Endomorphisms Universally of Trace Zero Via Certain Trace Relations

We provide a sufficient condition that ensures the nilpotency of endomorphisms universally of trace zero of Schur-finite objects in a category of homological type, i.e., a ?-linear ?-category with a tensor functor to super vector spaces. This generalizes previous results about finite-dimensional objects, in particular by Kimura in the category of motives. We also present some facts which suggest that this might be the best generalization possible of this line of proof. To get the result we prove an identity of trace relations on super vector spaces which has an independent interest in the field of combinatorics. Our main tool is Berele–Regev's theory of Hook Schur functions. We use their generalization of the classic Schur–Weyl duality to the “super” case, together with their factorization formula.     

A special class of tensor categories initiated by inverse braid monoids

In this paper, we introduce the concept of a wide tensor category which is a special class of a tensor category initiated by the inverse braid monoids recently investigated by Easdown and Lavers [The Inverse Braid Monoid, Adv. in Math. 186 (2004) 438-455].The inverse braid monoidsIB_n is an inverse monoid which behaves as the symmetric inverse semigroup so that the braid group B_n can be regarded as the braids acting in the symmetric group. In this paper, the structure of inverse braid monoids is explained by using the language of categories. A partial algebra category, which is a subcategory of the representative category of a bialgebra, is given as an example of wide tensor categories. In addition, some elementary properties of wide tensor categories are given. The main result is to show that for every strongly wide tensor category C, a strict wide tensor category C^str can be constructed and is wide tensor equivalent to C with a wide tensor equivalence F.As a generalization of the universality property of the braid category B, we also illustrate a wide tensor category through the discussion on the universality of the inverse braid category IB (see Theorem 3.3, 3.6 and Proposition 3.7).     

Unexpected properties of locally presentable categories

Locally presentable categories are precisely those complete categories which have a small dense subcategory. Every subcategory of a locally presentable category (i) has a small dense subcategory, (ii) if it is closed under limits, then it is locally presentable and (iii) if it is closed under colimits, then it is coreflective. For density, canonical colimits can be substituted by arbitrary colimits.The above results hold under the assumption of the set-theoretical Vopenka's Principle; in fact, each of them is logically equivalent to that principle.In Memory of Evelyn NelsonPresented by Ervin Fried.     

On Generation and Implicit Partial Operations in Locally Presentable Categories

In a locally presentable category seen as a concrete category of structures, we describe the subobjects (resp. the regular, strong subobjects) generated by a subset, first in terms of closure under certain types of implicit partial operations, and then in syntactic terms. This characterizes in particular the locally presentable categories in which various sorts of monos and epis coincide.     

Order-adjoint monads and injective objects

Given a monad T on whose functor factors through the category of ordered sets with left adjoint maps, the category of Kleisli monoids is defined as the category of monoids in the hom-sets of the Kleisli category of T. The Eilenberg-Moore category of T is shown to be strictly monadic over the category of Kleisli monoids. If the Kleisli category of T moreover forms an order-enriched category, then the monad induced by the new situation is Kock-Zöberlein. Injective objects in the category of Kleisli monoids with respect to the class of initial morphisms then characterize the objects of the Eilenberg-Moore category of T, a fact that allows us to recuperate a number of known results, and present some new ones.     

The wΔ-space problem

We address the following question: “Must every wΔ-space with a G_δ-diagonal be developable?” Consistently, the answer is “no.”Example. Assume CH. There is a zero-dimensional, scattered, locally compact, wΔ-space with a G_δ-diagonal which is not developable.For normal, locally compact spaces (or slightly weaker), the answer is “yes”.Theorem. If X is ω-sCWH, locally Lindelöf, wΔ-space with a G_δ-diagonal, then X is developable.     

Localization with Respect to Endomorphisms

Abstract. We introduce the class of localizable monoids. It contains inverse monoids. Then we define localizations of monoids with respect to localizable submonoids of their monoid of endomorphisms. These constructions can be applied to a category of left modules or to a category of A -rings. As a result, we are able to invert endomorphisms within the original category, unlike inversive localizations of Cohn's type which need a base change.     

Covariant Isotropy of Grothendieck Toposes and Extensive Categories

We provide an explicit characterization of the covariant isotropy group of any Grothendieck topos, i.e. the group of (extended) inner automorphisms of any sheaf over a small site. In order to do so, we first extend previous techniques for computing covariant isotropy from locally finitely presentable categories to locally presentable categories. As a consequence, we also obtain an explicit characterization of the centre of a Grothendieck topos, i.e. the automorphism group of its identity functor. We conclude by providing a more categorical approach to show that these characterizations also extend to any extensive category.

     

On Hölder Continuity of Quasiconformal Maps with VMO Dilatation

If the dilatation tensor or the matrix dilatation of a quasiconformal mapping $ f\!:\! G\to {\bf R} ^ n $ belongs to the space VMO of functions with vanishing mean oscillation, then f is locally Hölder continuous with every exponent f < 1.     

Sur le produit tensoriel des groupes affines

Let A and B be two commutative affine group schemes over a field. There exists an affine group A?B such that Hom(A?B,C)?Bil(A×B,C) for any affine group C. We use technics of the commutative algebraic groups theory, in order to compute these tensor products and to characterize “flat” groups in the unipotent case. The tensor product of commutative affine groups has most properties of the usual tensor product but it is not always associative. As an application we prove a structure theorem of the category of modules over some affine connected prosmooth rings.     

Deitmar’s versus To?n-Vaquié’s schemes over {\mathbb{F}_{1}}

Deitmar introduced schemes over ${\mathbb {F}_{1}}$ , the so-called “field with one element”, as certain spaces with an attached sheaf of monoids, generalizing the definition of schemes as ringed spaces. On the other hand, To?n and Vaquié defined them as particular Zariski sheaves over the opposite category of monoids, generalizing the definition of schemes as functors of points. We show the equivalence between Deitmar’s and To?n-Vaquiés notions and establish an analog of the classical case of schemes over ${\mathbb {Z}}$ . This result has been assumed by the leading experts on ${\mathbb {F}_{1}}$ , but no proof was given. During the proof, we also conclude some new basic results on commutative algebra of monoids, such as a characterization of local flat epimorphisms and of flat epimorphisms of finite presentation. We also inspect the base-change functors from the category of schemes over ${\mathbb {F}_{1}}$ to the category of schemes over ${\mathbb {Z}}$ .     

Accessible aspects of 2-category theory

Categorical structures and their pseudomaps rarely form locally presentable 2-categories in the sense of Cat-enriched category theory. However, we show that if the categorical structure in question is sufficiently weak (such as the structure of monoidal, but not strict monoidal, categories) then the 2-category in question is accessible. Furthermore, we explore the flexible limits that such 2-categories possess and their interaction with filtered colimits.