Alphabeta Math
PropositionStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passaudited 2026-09-27
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Additive functors preserve chosen Gaussian cancellations

Statement

Let F:A→B be an additive functor between additive categories, let X∙ be a cochain complex in A with a pivot decomposition at degree n, Schur complement dˉ=a−bφ−1c, candidate reduction Xˉ∙ and two-term complex K as in An invertible cochain differential block and its candidate reduction and Gaussian elimination splits a contractible two-term complex, and let (p,ı,h) be the strong deformation retract data of Explicit strong deformation retract from Gaussian cancellation.

  1. Complexes and pivots. F(X∙), with differentials F(dn), is a cochain complex in B; the pivot F(φ) is invertible with inverse F(φ−1); and with respect to the biproduct decompositions F(Xn)=F(A)⊕F(U), F(Xn+1)=F(B)⊕F(V) whose structure maps are the F-images of those of X∙, the differential F(dn) has the entrywise image matrix (F(a)F(b)F(c)F(φ)).
  2. The corresponding cancellation. The reduction of F(X∙) at the pivot F(φ) is F(Xˉ∙): its objects and neighbouring arrows are the F-images of those of Xˉ∙, and its differential in degree n is the Schur complement F(a)−F(b)F(φ)−1F(c)=F(dˉ).
  3. Retract data. The images F(p),F(ı),F(h) satisfy F(p)F(ı)=1, 1−F(ı)F(p)=F(d)F(h)+F(h)F(d), F(p)F(h)=0, F(h)F(ı)=0 and F(h)2=0, so they are strong deformation retract data of F(X∙) onto F(Xˉ∙). Moreover F(K) is the two-term complex F(U)→F(φ)F(V) with vanishing neighbouring terms, contractible via F(φ−1), and F(T),F(T−1) remain mutually inverse cochain isomorphisms between F(X∙) and F(Xˉ∙⊕K).
  4. Scope. Clauses 1 to 3 use only additivity: no exactness of F is assumed or needed. If B is abelian, the image retract maps induce inverse isomorphisms on the homology of F(X∙) and F(Xˉ∙), by Gaussian cancellation preserves homotopy type and abelian-category homology. No comparison of F(Hn(X)) with Hn(F(X)) is asserted; these expressions both make sense when A and B are abelian, but comparing them is a separate question about commuting F with homology.

Facts & Assumptions

Given: An additive functor F:A→B between additive categories, a cochain complex X∙ in A with the pivot decomposition at degree n, its reduction Xˉ∙, the two-term complex K, the chain isomorphism T, and the explicit cochain maps p,ı and homotopy h of the strong deformation retract.

[L1]

p,ı are cochain maps and h has degree −1, with pı=1Xˉ∙, 1X∙−ıp=dh+hd, ph=0, hı=0 and h2=0 (Explicit strong deformation retract from Gaussian cancellation).

[L2]

The decomposition Xn=A⊕U, Xn+1=B⊕V has φ:U→V invertible, dn=(abcφ), dn−1=(p;q), dn+1=(r s), and the candidate reduction replaces degrees n,n+1 by A,B with dn replaced by dˉ=a−bφ−1c and neighbouring arrows p and r (An invertible cochain differential block and its candidate reduction).

[L3]

T:X∙→Xˉ∙⊕K is an isomorphism of cochain complexes with inverse T−1, where K has Kn=U, Kn+1=V, vanishing terms elsewhere and differential φ, and K is contractible with contracting homotopy φ−1 in degree n+1 (Gaussian elimination splits a contractible two-term complex).

[L4]

An additive functor preserves composition and identities, and its induced maps on hom-groups are homomorphisms: F(f+g)=F(f)+F(g) and hence F(−f)=−F(f) (Additive functor).

[L5]

An additive functor preserves finite biproducts, so the F-images of the injections and projections of a finite biproduct exhibit F(A⊕U) as a biproduct F(A)⊕F(U) with the same identity-sum relations; it also preserves zero morphisms; and composition of morphisms between finite biproducts is matrix multiplication (An additive functor preserves finite biproducts, An additive functor preserves zero morphisms, Composition of morphisms between finite biproducts is matrix multiplication, Complexes, homotopies and contractibility in an additive category).

[L6]

In an abelian category, the maps of a Gaussian strong deformation retract induce mutually inverse maps on every homology object of the reindexed chain complexes (Gaussian cancellation preserves homotopy type and abelian-category homology).

Proof

technique · direct
1.1

Complex and pivot. Since dn+1dn=0 in X∙, [L4] gives F(dn+1)F(dn)=F(dn+1dn)=F(0Xn,Xn+2), which is the zero morphism by [L5]; thus F(X∙) is a cochain complex. Likewise F(φ)F(φ−1)=F(φφ−1)=F(1V)=1F(V) and F(φ−1)F(φ)=1F(U), so F(φ) is invertible with the displayed inverse.

L4L5algebra
1.2

Image matrices. By [L5] the F-images of the injections and projections of Xn=A⊕U and Xn+1=B⊕V exhibit F(Xn) as F(A)⊕F(U) and F(Xn+1) as F(B)⊕F(V). Writing dn=iBapA+iBbpU+iVcpA+iVφpU with the biproduct structure maps [L2], additivity of F on hom-groups, preservation of composition and the biproduct relations give F(dn)=F(iB)F(a)F(pA)+F(iB)F(b)F(pU)+F(iV)F(c)F(pA)+F(iV)F(φ)F(pU), whose matrix with respect to the image decompositions is (F(a)F(b)F(c)F(φ)) by the matrix convention of [L5]. The same computation applies to dn−1 and dn+1, giving the image neighbouring components F(p),F(q),F(r),F(s).

L4L5algebra
1.3

Retract identities are preserved. Applying [L4] to the identities of [L1] and using [L5] for the zero morphisms: F(p)F(ı)=F(pı)=F(1Xˉ∙)=1F(Xˉ∙); F(d)F(h)+F(h)F(d)=F(dh+hd)=F(1X∙−ıp)=1F(X∙)−F(ı)F(p); F(p)F(h)=F(ph)=F(0)=0; F(h)F(ı)=F(hı)=0; and F(h)2=F(h2)=F(0)=0. Since p,ı are cochain maps, F(p),F(ı) are cochain maps by [L4].

L1L4L5algebra
2.1

The contractible summand and the isomorphism. By [L3] and [L4], F(T)F(T−1)=F(TT−1)=F(1)=1 and F(T−1)F(T)=1, so F(T) is an isomorphism of complexes with inverse F(T−1); and F(K) has objects F(U),F(V) in degrees n,n+1, vanishing terms elsewhere with zero differentials, and differential F(φ), with F(φ)F(φ−1)=1F(V) and F(φ−1)F(φ)=1F(U) from step 1.1, so F(φ−1) is a contracting homotopy for F(K).

L3L4L5step 1.1algebra
2.2

The reduction of F(X∙) is F(Xˉ∙). By step 1.2 the reduction problem for F(X∙) in degrees n,n+1 is the image matrix (F(a)F(b)F(c)F(φ)) with pivot F(φ), and by step 1.1 that pivot is invertible; [L4] gives F(a)−F(b)F(φ−1)F(c)=F(a−bφ−1c)=F(dˉ), and applying F to the remaining data of [L2] gives objects F(A),F(B) in degrees n,n+1, neighbouring arrows F(p),F(r) and the unchanged images of the outside objects and arrows. Hence the candidate reduction of F(X∙) at this pivot is exactly F(Xˉ∙), its differential in degree n being the image F(dˉ) of the Schur complement.

L2L4step 1.1step 1.2algebra
3.1

Conclusion. Step 1.1 shows that F(X∙) is a complex with invertible pivot F(φ), step 1.2 computes the image matrices, and step 2.2 identifies the reduction of F(X∙) with F(Xˉ∙), which is clause 2 and the matrix assertion of clause 1. Step 1.3 verifies all five strong deformation retract identities for F(p),F(ı),F(h), and step 2.1 shows that F(K) is contractible via F(φ−1) and that F(T) is an isomorphism, which is clause 3. Only additivity, preservation of finite biproducts and preservation of zero morphisms are used, so no exactness hypothesis enters; if B is abelian, [L6] applied to the image cancellation identified in step 2.2 gives inverse homology maps. This compares the homology of the two image complexes, not the image under F of a homology object in A. ∎

L6step 2.1step 2.2step 1.3step 1.2step 1.1algebra

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

29 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources