Documentation

Toric.Hopf.Diagonalisable

Diagonalizable bialgebras #

We define diagonalizable bialgebras (over a commutative ring R) as those that are isomorphic to a group algebra over R.

We then prove that any diagonalizable bialgebra is spanned by its group-like elements.

If the base ring is a domain, we prove the converse: any bialgebra spanned by its group-like elements is diagonalizable. The idea is that, if A is a bialgebra over R and G is its set of group-like elements, then G is a group (for the multiplication of A), so we get a morphism of algebras from R[G] to A, which is actually a morphism of bialgebras. This morphism is surjective by assumption, and, if R is a domain, it is also injective because group-like elements are linearly independent.

Note that the last result is false in general.

class Bialgebra.IsDiagonalisable (R : Type u_1) (A : Type u) [CommRing R] [Semiring A] [Bialgebra R A] :

A bialgebra is called diagonalisable if it is isomorphic to a group algebra.

Instances
    theorem Bialgebra.isDiagonalisable_iff (R : Type u_1) (A : Type u) [CommRing R] [Semiring A] [Bialgebra R A] :
    IsDiagonalisable R A ∃ (G : Type u) (x : Group G), Nonempty (A ≃ₐc[R] MonoidAlgebra R G)

    A group algebra is diagonalisable.

    theorem Bialgebra.IsDiagonalisable.ofBialgEquiv {R : Type u_1} {A : Type u} [CommRing R] [Semiring A] [Bialgebra R A] {B : Type u_3} [Semiring B] [Bialgebra R B] [hA : IsDiagonalisable R A] (e : A ≃ₐc[R] B) :

    A diagonalisable bialgebra is generated by its group-like elements.

    A Hopf algebra over a domain that is generated by its group-like elements is isomorphic to the group algebra on its group-like elements.

    A Hopf algebra over a domain that is generated by its group-like elements is diagonalisable.

    This is also true over a commutative ring, but with a more complicated proof.

    A Hopf algebra over a domain is diagonalizable if and only if it is generated by its group-like elements.

    This is also true over a commutative ring, but with a more complicated proof.