Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generated
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.

Kernels, images, quotients, and matrix coefficients of comodules

Statement

Let C be a coalgebra over a field k. The regular pair (C,Δ) is a right C-comodule. If f:M→N is a right-comodule map, then ker⁡f and im⁡f inherit right-comodule structures, the quotient M/ker⁡f is a right comodule, and the induced map M/ker⁡f→im⁡f is a comodule isomorphism, assuming AC. For a finite-dimensional right comodule M with ordered basis v1,…,vn, write ρ(vj)=∑i=1nvi⊗cij. Then Δ(cij)=∑ℓ=1nciℓ⊗cℓj,ε(cij)=δij. Conversely, any n2 elements cij∈C satisfying these two identities define a right-comodule structure on the vector space with basis v1,…,vn by the displayed formula.

Facts & Assumptions

Given: A field k, a coalgebra C with maps Δ and ε, right C-comodules (M,ρM) and (N,ρN), a right-comodule map f:M→N, and, for the coefficient clause, a finite-dimensional right comodule with a fixed ordered basis.

[F1]

A right comodule has a linear map ρ:X→X⊗C satisfying (ρ⊗id⁡)ρ=(id⁡⊗Δ)ρ and (id⁡⊗ε)ρ=id⁡; a map h:X→Y is a comodule map when ρYh=(h⊗id⁡)ρX. (Right comodules, comodule maps, and subcomodules)

[F2]

A coalgebra has coassociative comultiplication Δ:C→C⊗C and a counit ε:C→k for which both counit composites are the identity through the tensor-unit isomorphisms; coalgebra maps preserve both structure maps. (Coalgebra and coalgebra map over a field)

[F3]

The tensor products are over k, every tensor is a finite sum of elementary tensors, and tensor powers are re-associated only through the stated associators. (Scalars, tensor powers, the empty tensor, opposite algebras and finite sums, Associator naturality, pentagon, unit triangle and symmetry hexagons on elementary tensors)

[F4]

Under AC, tensoring an injective linear map by id⁡V remains injective; for quotient maps p:X→X/U and q:Y→Y/Z, ker⁡(p⊗q) is the image of U⊗Y+X⊗Z in X⊗Y. (Tensoring injections and the kernel of a tensor product of quotient maps over a field)

[F5]

Tensor maps act on elementary tensors by (g⊗h)(x⊗y)=g(x)⊗h(y) and preserve identities and composition. (Module homomorphisms induce tensor-product homomorphisms functorially)

[F6]

ker⁡f={x:f(x)=0} and im⁡f={f(x):x∈M}; these are subspaces for a linear map. (Kernel and image of a linear map, The kernel and image are linear subspaces, and a linear map is injective if and only if its kernel is trivial)

[F7]

If K is a submodule, M/K consists of cosets m+K and the quotient map is q(m)=m+K. (Submodule of a module, Quotient module M/N with scalar multiplication on additive cosets)

[A1]

AC holds, so the tensor-injection and tensor-kernel conclusions in [F4] apply to arbitrary vector spaces. (The Axiom of Choice)

Proof

Given: The objects and map in the statement, with AC for the arbitrary-dimensional kernel, image, and quotient claims.

1.1givenF1F2

For the regular object take ρC=Δ. The coalgebra coassociativity identity is exactly (ρC⊗id⁡C)ρC=(id⁡C⊗Δ)ρC, and its two counit identities give (id⁡C⊗ε)ρC=id⁡C (and the other counit identity), so (C,Δ) is a right comodule by the coalgebra definition [F2].

1.2givenF3F5F8algebrachoose

Let M have ordered basis v1,…,vn, and write λi(x) for the unique coordinates of x from [F8]. For a,b∈k and x,y∈M, the coordinate list of ax+by is (aλi(x)+bλi(y))i by the vector-space laws; uniqueness in [F8] therefore gives λi(ax+by)=aλi(x)+bλi(y), so each coordinate function λi:M→k is linear. For each j, write ρM(vj)=∑sxs⊗ds as a finite tensor by [F3], expand each xs=∑iλi(xs)vi, and collect cij:=∑sλi(xs)ds. Applying λr⊗id⁡C recovers crj, so the expansion ρM(vj)=∑ivi⊗cij is unique.

1.3givenA1F4F6F7F5algebra

Let K=ker⁡f, I=im⁡f, and let q:M→M/K be the quotient. The map fˉ:M/K→N, fˉ(m+K)=f(m), is well-defined and injective: equal cosets have difference in K, and fˉ(m+K)=0 implies m∈K. It is linear because for a,b∈k and cosets represented by m,n∈M, fˉ(a(m+K)+b(n+K))=f(am+bn)=af(m)+bf(n). Thus f=fˉq, so functoriality [F5] gives f⊗id⁡C=(fˉ⊗id⁡C)(q⊗id⁡C). By [A1] and [F4], fˉ⊗id⁡C is injective, while the kernel formula in [F4] with U=K and Z=0 gives ker⁡(q⊗id⁡C)=im⁡(K⊗C). Therefore ker⁡(f⊗id⁡C)=im⁡(K⊗C→M⊗C). The same tensor-injection result makes the inclusion K⊗C→M⊗C injective.

2.1givenF1F8step 1.2algebra

Apply coassociativity to vj and use the expansion in 1.2 to get ∑i,ℓvi⊗ciℓ⊗cℓj=∑ivi⊗Δ(cij). Applying λr⊗id⁡C⊗id⁡C gives Δ(crj)=∑ℓcrℓ⊗cℓj. Using the counit equation (id⁡M⊗ε)ρM(vj)=vj under the tensor-unit identification and then applying λr gives ε(crj)=δrj.

2.2step 1.3F1F4F5algebra

For x∈K, the comodule-map equation gives (f⊗id⁡)ρM(x)=ρN(f(x))=0. By step 1.3, ρM(x) has a unique preimage in K⊗C under the injective map i⊗id⁡, where i:K↪M; define ρK(x) to be that preimage. This is linear because it is the inverse of a linear isomorphism onto its image. The identity (i⊗id⁡)ρK=ρMi and the injectivity of i⊗id⁡⊗id⁡ show that coassociativity of ρK follows from coassociativity of ρM; injectivity of i and the same identity show its counit law. Hence K is a subcomodule and i is a comodule map.

3.1givenF1F2F8step 1.2step 2.1algebra

Conversely, suppose the displayed matrix identities hold. Define ρ on the basis by ρ(vj)=∑ivi⊗cij and extend linearly using the unique coordinates from [F8]; the coordinate functionals are linear by step 1.2. For each basis vector, (ρ⊗id⁡)ρ(vj)=∑i,ℓvi⊗ciℓ⊗cℓj=∑ivi⊗Δ(cij)=(id⁡⊗Δ)ρ(vj) by the coproduct identity; also (id⁡⊗ε)ρ(vj)=∑iviδij=vj. The maps in each identity are linear and the basis spans M, so both identities hold on every vector; hence this is a right comodule.

3.2step 2.2F1F5F6algebra

Factor f as M→qII→jN, where I=im⁡f is a subspace by [F6], qI(m):=f(m) with codomain restricted to I is the linear corestriction of f, and j is inclusion. The map qI is onto by the definition of I. Define ρI(qI(m)):=(qI⊗id⁡)ρM(m). If qI(m)=qI(m′), then m−m′∈K; by 2.2, ρM(m−m′) lies in the image of K⊗C, which qI⊗id⁡ kills, so the formula is well-defined. It is linear and satisfies ρIqI=(qI⊗id⁡)ρM. Precomposing the coassociativity identity for ρI with the surjection qI gives (qI⊗id⁡⊗id⁡)(ρM⊗id⁡)ρM=(qI⊗id⁡⊗id⁡)(id⁡⊗Δ)ρM, so it holds by coassociativity on M and surjectivity of qI. Likewise, precomposing its counit identity with qI gives qI(id⁡⊗ε)ρM=qI, which holds by counitality on M; surjectivity gives the identity on I. Thus ρI is a coaction. Since f=jqI is a comodule map, ρNjqI=(f⊗id⁡)ρM=(j⊗id⁡)(qI⊗id⁡)ρM=(j⊗id⁡)ρIqI; surjectivity of qI proves ρNj=(j⊗id⁡)ρI, so j is a comodule map.

4.1step 2.2step 3.2F1F5F7algebra

Let i:J0→M be an injective comodule map and set J=i(J0). By step 3.2, J has the image coaction making its inclusion a comodule map. Let π:M→M/J be the quotient and set ρˉ(m+J):=(π⊗id⁡)ρM(m). If m is replaced by m+j, the difference is (π⊗id⁡)ρM(j)=0 because ρM(j) is the image of a tensor in J⊗C; hence ρˉ is well-defined. It is linear and ρˉπ=(π⊗id⁡)ρM. Precomposing its coassociativity identity with the surjection π gives (π⊗id⁡⊗id⁡)(ρM⊗id⁡)ρM=(π⊗id⁡⊗id⁡)(id⁡⊗Δ)ρM, so it holds by coassociativity on M and surjectivity of π. Precomposing its counit identity with π gives π(id⁡⊗ε)ρM=π, which follows by counitality on M; surjectivity gives the identity on M/J. Thus the quotient is a comodule and π is a comodule map.

5.1step 3.2step 4.1F5F6F7algebra

The map f~:M/K→I, m+K↦f(m), is the same induced map as fˉ with codomain restricted to I. It is injective by the definition of K and onto by the definition of I, hence is a vector-space isomorphism. The quotient and image coaction formulas in steps 3.2 and 4.1 give ρIf~=(f~⊗id⁡)ρˉ. Since f~ and its tensor map are isomorphisms, composing this identity with their inverses shows f~−1 is also a comodule map.

6.1step 1.1step 1.2step 2.1step 3.1step 1.3step 2.2step 3.2step 4.1step 5.1∎

Step 1.1 establishes the regular comodule; steps 1.2, 2.1, and 3.1 prove both directions of the finite coefficient-matrix characterization. Steps 1.3, 2.2, 3.2, 4.1, and 5.1 establish the kernel, image, quotient, and first-isomorphism claims under AC.

Depends on

Used by

Cited to discharge well-definedness by Right comodules, comodule maps, and subcomodules.

Dependency tree · two levels

46 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