Documentation

TauCeti.Algebra.Coalgebra.Comodule.LinearHom

Linear Hom comodules #

For comodules M and N over a Hopf algebra, with M finite projective, Hom_R(M,N) is a comodule by transport from the diagonal tensor comodule M* ⊗ N. On algebra-valued points its action is f ↦ g_N ∘ f ∘ g_M⁻¹. In particular, taking N = M gives the conjugation representation on endomorphisms, used to construct representations with prescribed normal kernels. For a commutative Hopf algebra, the fixed vectors are exactly the comodule morphisms.

The order M* ⊗ N is part of this construction. Over a noncommutative Hopf algebra, swapping it to N ⊗ M* need not be colinear, and the identity endomorphism need not be fixed in the former. The fixed-morphism characterization below therefore retains commutative coefficients.

The construction uses Comodule.dual, Comodule.tensor, Comodule.Transport and Mathlib's dualTensorHomEquiv; the point formula uses the existing inverse-point evaluation identity.

References #

@[implicit_reducible]
noncomputable def TauCeti.Comodule.linearHom {R : Type u} {H : Type v} {M : Type w} {N : Type x} [CommSemiring R] [AddCommMonoid M] [Module R M] [AddCommMonoid N] [Module R N] [Module.Finite R M] [Module.Projective R M] [Semiring H] [HopfAlgebra R H] [Comodule R H M] [Comodule R H N] :
Comodule R H (M →ₗ[R] N)

The comodule of linear maps from a finite projective comodule to an arbitrary comodule, with the contragredient action on the source. This is an explicit structure, not a global instance, so different coactions on the same linear Hom space can coexist.

Equations
Instances For
    @[simp]
    theorem TauCeti.Comodule.linearHom_coact {R : Type u} {H : Type v} {M : Type w} {N : Type x} [CommSemiring R] [AddCommMonoid M] [Module R M] [AddCommMonoid N] [Module R N] [Module.Finite R M] [Module.Projective R M] [Semiring H] [HopfAlgebra R H] [Comodule R H M] [Comodule R H N] (f : M →ₗ[R] N) :

    The Hom coaction is transported from the tensor product of the dual and the target.

    @[simp]

    Under scalar extension of linear Hom, an algebra-valued point acts by conjugating with its actions on the target and source.

    A scalar-extended linear map is fixed by a point in the Hom comodule exactly when it intertwines the point actions on its source and target.

    @[simp]

    The fixed vectors in the linear Hom comodule are exactly the colinear linear maps. This criterion includes nonreduced Hopf algebras and needs no hypothesis on geometric points.

    def TauCeti.Comodule.fixedLinearHomEquiv {R : Type u} {H : Type v} {M : Type w} {N : Type x} [CommSemiring R] [AddCommMonoid M] [Module R M] [AddCommMonoid N] [Module R N] [Module.Finite R M] [Module.Projective R M] [CommSemiring H] [HopfAlgebra R H] [Comodule R H M] [Comodule R H N] :
    ↥(fixedSubcomodule R H (M →ₗ[R] N)) ≃ₗ[R] Hom R H M N

    Comodule morphisms are the fixed vectors of the linear Hom comodule.

    Equations
    • One or more equations did not get rendered due to their size.
    Instances For
      @[simp]

      The fixed-Hom equivalence preserves the underlying linear map.

      @[simp]
      theorem TauCeti.Comodule.fixedLinearHomEquiv_symm_coe {R : Type u} {H : Type v} {M : Type w} {N : Type x} [CommSemiring R] [AddCommMonoid M] [Module R M] [AddCommMonoid N] [Module R N] [Module.Finite R M] [Module.Projective R M] [CommSemiring H] [HopfAlgebra R H] [Comodule R H M] [Comodule R H N] (f : Hom R H M N) :

      The inverse fixed-Hom equivalence preserves the underlying linear map.