Hom vanishing into an almost-split kernel kills extensions #
theorem
MagnitudeConjecture.CategoryTheory.splitEpi_of_hom_to_almostSplit_kernel_zero
{C : Type u}
[CategoryTheory.Category.{v, u} C]
[CategoryTheory.Abelian C]
{S T : CategoryTheory.ShortComplex C}
(hS : S.ShortExact)
(hT : T.ShortExact)
(hTg : QuotientSubmoduleEquidistribution.IsRightAlmostSplit T.g)
(e : S.X₃ ≅ T.X₃)
(hHom : ∀ (f : S.X₁ ⟶ T.X₁), f = 0)
:
CategoryTheory.IsSplitEpi S.g
A short exact sequence splits if its kernel has no maps to the kernel of a right almost-split sequence with the same endpoint.
theorem
MagnitudeConjecture.CategoryTheory.extOne_eq_zero_of_hom_to_almostSplit_kernel_zero
{C : Type u}
[CategoryTheory.Category.{v, u} C]
[CategoryTheory.Abelian C]
[CategoryTheory.HasExt C]
[CategoryTheory.EnoughProjectives C]
{T : CategoryTheory.ShortComplex C}
(hT : T.ShortExact)
(hTg : QuotientSubmoduleEquidistribution.IsRightAlmostSplit T.g)
(K : C)
(hHom : ∀ (f : K ⟶ T.X₁), f = 0)
(xi : CategoryTheory.Abelian.Ext T.X₃ K 1)
:
xi = 0
The Hom-vanishing consequence of AR duality follows directly from the factorization property and realization of degree-one extensions.
theorem
MagnitudeConjecture.CategoryTheory.hom_lift_of_extOne_eq_zero
{C : Type u}
[CategoryTheory.Category.{v, u} C]
[CategoryTheory.Abelian C]
[CategoryTheory.HasExt C]
{S : CategoryTheory.ShortComplex C}
(hS : S.ShortExact)
(Q : C)
(hExt : ∀ (xi : CategoryTheory.Abelian.Ext Q S.X₁ 1), xi = 0)
(f : Q ⟶ S.X₃)
:
∃ (g : Q ⟶ S.X₂), CategoryTheory.CategoryStruct.comp g S.g = f
Ext vanishing lifts every map through the epimorphism of a short exact sequence.