Documentation

Mathlib.Algebra.Category.MonCat.FilteredColimits

The forgetful functor from (commutative) (additive) monoids preserves filtered colimits. #

Forgetful functors from algebraic categories usually don't preserve colimits. However, they tend to preserve filtered colimits.

In this file, we start with a small filtered category J and a functor F : J ⥤ MonCat. We then construct a monoid structure on the colimit of F ⋙ forget MonCat (in Type), thereby showing that the forgetful functor forget MonCat preserves filtered colimits. Similarly for AddMonCat, CommMonCat and AddCommMonCat.

The colimit of F ⋙ forget AddMon in the category of types. In the following, we will construct an additive monoid structure on M.

Instances For
    @[inline, reducible]

    The colimit of F ⋙ forget MonCat in the category of types. In the following, we will construct a monoid structure on M.

    Instances For

      The canonical projection into the colimit, as a quotient type.

      Instances For
        @[inline, reducible]

        The canonical projection into the colimit, as a quotient type.

        Instances For
          theorem AddMonCat.FilteredColimits.M.mk_eq {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J AddMonCat) (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) (h : k f g, ↑(F.map f) x.snd = ↑(F.map g) y.snd) :
          theorem MonCat.FilteredColimits.M.mk_eq {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J MonCat) (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) (h : k f g, ↑(F.map f) x.snd = ↑(F.map g) y.snd) :

          As J is nonempty, we can pick an arbitrary object j₀ : J. We use this object to define the "zero" in the colimit as the equivalence class of ⟨j₀, 0 : F.obj j₀⟩.

          As J is nonempty, we can pick an arbitrary object j₀ : J. We use this object to define the "one" in the colimit as the equivalence class of ⟨j₀, 1 : F.obj j₀⟩.

          The definition of the "zero" in the colimit is independent of the chosen object of J. In particular, this lemma allows us to "unfold" the definition of colimit_zero at a custom chosen object j.

          The definition of the "one" in the colimit is independent of the chosen object of J. In particular, this lemma allows us to "unfold" the definition of colimit_one at a custom chosen object j.

          noncomputable def AddMonCat.FilteredColimits.colimitAddAux {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J AddMonCat) [CategoryTheory.IsFiltered J] (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) :

          The "unlifted" version of addition in the colimit. To add two dependent pairs ⟨j₁, x⟩ and ⟨j₂, y⟩, we pass to a common successor of j₁ and j₂ (given by IsFiltered.max) and add them there.

          Instances For
            noncomputable def MonCat.FilteredColimits.colimitMulAux {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J MonCat) [CategoryTheory.IsFiltered J] (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) :

            The "unlifted" version of multiplication in the colimit. To multiply two dependent pairs ⟨j₁, x⟩ and ⟨j₂, y⟩, we pass to a common successor of j₁ and j₂ (given by IsFiltered.max) and multiply them there.

            Instances For

              Addition in the colimit is well-defined in the left argument.

              Multiplication in the colimit is well-defined in the left argument.

              Addition in the colimit is well-defined in the right argument.

              Multiplication in the colimit is well-defined in the right argument.

              theorem AddMonCat.FilteredColimits.colimit_add_mk_eq {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J AddMonCat) [CategoryTheory.IsFiltered J] (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) (k : J) (f : x.fst k) (g : y.fst k) :
              AddMonCat.FilteredColimits.M.mk F x + AddMonCat.FilteredColimits.M.mk F y = AddMonCat.FilteredColimits.M.mk F { fst := k, snd := ↑(F.map f) x.snd + ↑(F.map g) y.snd }

              Addition in the colimit is independent of the chosen "maximum" in the filtered category. In particular, this lemma allows us to "unfold" the definition of the addition of x and y, using a custom object k and morphisms f : x.1 ⟶ k and g : y.1 ⟶ k.

              theorem MonCat.FilteredColimits.colimit_mul_mk_eq {J : Type v} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J MonCat) [CategoryTheory.IsFiltered J] (x : (j : J) × ↑(F.obj j)) (y : (j : J) × ↑(F.obj j)) (k : J) (f : x.fst k) (g : y.fst k) :
              MonCat.FilteredColimits.M.mk F x * MonCat.FilteredColimits.M.mk F y = MonCat.FilteredColimits.M.mk F { fst := k, snd := ↑(F.map f) x.snd * ↑(F.map g) y.snd }

              Multiplication in the colimit is independent of the chosen "maximum" in the filtered category. In particular, this lemma allows us to "unfold" the definition of the multiplication of x and y, using a custom object k and morphisms f : x.1 ⟶ k and g : y.1 ⟶ k.

              The bundled additive monoid giving the filtered colimit of a diagram.

              Instances For

                The bundled monoid giving the filtered colimit of a diagram.

                Instances For

                  The additive monoid homomorphism from a given additive monoid in the diagram to the colimit additive monoid.

                  Instances For

                    The monoid homomorphism from a given monoid in the diagram to the colimit monoid.

                    Instances For

                      The cocone over the proposed colimit additive monoid.

                      Instances For

                        The cocone over the proposed colimit monoid.

                        Instances For

                          Given a cocone t of F, the induced additive monoid homomorphism from the colimit to the cocone point. As a function, this is simply given by the induced map of the corresponding cocone in Type. The only thing left to see is that it is an additive monoid homomorphism.

                          Instances For
                            theorem AddMonCat.FilteredColimits.colimitDesc.proof_2 {J : Type u_2} [CategoryTheory.SmallCategory J] (F : CategoryTheory.Functor J AddMonCat) [CategoryTheory.IsFiltered J] (t : CategoryTheory.Limits.Cocone F) (x : ↑(AddMonCat.FilteredColimits.colimit F)) (y : ↑(AddMonCat.FilteredColimits.colimit F)) :
                            ZeroHom.toFun { toFun := CategoryTheory.Limits.IsColimit.desc (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t), map_zero' := (_ : CategoryTheory.Limits.IsColimit.desc J inst✝ TypeMax CategoryTheory.types (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat)) (CategoryTheory.Limits.Types.colimitCocone (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t) 0 = 0) } (x + y) = ZeroHom.toFun { toFun := CategoryTheory.Limits.IsColimit.desc (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t), map_zero' := (_ : CategoryTheory.Limits.IsColimit.desc J inst✝ TypeMax CategoryTheory.types (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat)) (CategoryTheory.Limits.Types.colimitCocone (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t) 0 = 0) } x + ZeroHom.toFun { toFun := CategoryTheory.Limits.IsColimit.desc (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t), map_zero' := (_ : CategoryTheory.Limits.IsColimit.desc J inst✝ TypeMax CategoryTheory.types (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat)) (CategoryTheory.Limits.Types.colimitCocone (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) (CategoryTheory.Limits.Types.colimitCoconeIsColimit (CategoryTheory.Functor.comp F (CategoryTheory.forget AddMonCat))) ((CategoryTheory.forget AddMonCat).mapCocone t) 0 = 0) } y

                            Given a cocone t of F, the induced monoid homomorphism from the colimit to the cocone point. As a function, this is simply given by the induced map of the corresponding cocone in Type. The only thing left to see is that it is a monoid homomorphism.

                            Instances For

                              The colimit of F ⋙ forget₂ AddCommMonCat AddMonCat in the category AddMonCat. In the following, we will show that this has the structure of a commutative additive monoid.

                              Instances For
                                @[inline, reducible]

                                The colimit of F ⋙ forget₂ CommMonCat MonCat in the category MonCat. In the following, we will show that this has the structure of a commutative monoid.

                                Instances For

                                  The bundled additive commutative monoid giving the filtered colimit of a diagram.

                                  Instances For

                                    The bundled commutative monoid giving the filtered colimit of a diagram.

                                    Instances For

                                      The cocone over the proposed colimit additive commutative monoid.

                                      Instances For

                                        The cocone over the proposed colimit commutative monoid.

                                        Instances For