Alphabeta Math
DefinitionDefinition: Literature-sourcedProof: AI-adaptedPipeline-generatedjudge pass (gpt-5.6-terra)audited 2026-09-09
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.

Low-degree transgression for a group extension

Definition

Assume AC. For 1→N→G→πQ→1, a left G-module A and [d]∈H1(N,A)Q, put Dd={(d(n),n):n∈N}≤A⋊G and Ld=NA⋊G(Dd). Define Tra⁡[d] to be the class of 0⟶AN⟶Ld/Dd⟶Q⟶1. It is a well-defined homomorphism to normalized bar H2(Q,AN). Explicitly choose normalized α:Q→G and η:Q→A with (α(q)⋅d−d)(n)=nη(q)−η(q); put f(q,r)=α(q)α(r)α(qr)−1. Then Tra is represented by F(q,r)=η(q)+α(q)η(r)−f(q,r)η(qr)−d(f(q,r)). This fixes the DHW sign convention. The derived interpretation of bar cohomology retains its comparison convention.

Facts & Assumptions

Given: AC, the extension, A, and a Q-invariant crossed-map class [d].

[F1]

The conjugation action on crossed-map classes factors through Q (Degree-one maps and the quotient action are well-defined).

[F2]

Normalized factor sets classify the extensions with fixed action (Bar two-cocycles classify abelian-kernel extensions).

[F3]

AC chooses elements of arbitrary nonempty indexed families (The Axiom of Choice).

Proof

1.1F1givenalgebra

The crossed identity makes Dd a subgroup mapping isomorphically onto N. In A⋊G, conjugating (d(g−1ng),g−1ng) by (a,g) gives (a+gd(g−1ng)−na,n). Thus (a,g)∈Ld exactly when (g⋅d−d)(n)=na−a for every n. By invariance in F1 there exists such an a for each g, so Ld→G is onto. Setting g=1 shows Ld∩A=AN. Its elements over N are precisely ANDd, since division by the unique D_d element over the same n lies in this intersection. Consequently the quotient extension is exact; its action on AN is a↦ga, the quotient Q-action.

2.1F3step 1.1algebra

Apply AC to the fibers of G→Q, normalizing α(1)=1. For each q the set of a solving the normalizer equation for α(q) is nonempty by step 1.1. Apply AC to these Q-indexed sets and normalize η(1)=0, which is a solution. Then lq=(η(q),α(q))∈Ld. For f=f(q,r), direct multiplication gives lqlr=(F(q,r),1)(d(f),f)lqr with exactly the F printed in the Definition. All factors except possibly (F,1) are in Ld, so it too is, and step 1.1 gives F∈AN.

2.2F2step 1.1algebra

If db=d+δb for a fixed b in A, then Ddb=(−b,1)Dd(−b,1)−1 by the semidirect multiplication. Conjugation by (-b,1) therefore identifies their normalizers and quotients, fixes every element of AN, and is the identity on Q. Hence the extension class depends only on [d].

3.1F2step 2.1algebra

In the quotient Ld/Dd, the elements lˉq form a normalized section and satisfy lˉqlˉr=i(F(q,r))lˉqr. Associativity yields F(q,r)+F(qr,s)=qF(r,s)+F(q,rs) by comparing the two triple products. Also F(1,q)=F(q,1)=0. Thus F is a normalized AN-valued cocycle representing the extension by F2. Different alpha or eta give another normalized section of the same extension; their unique difference b:Q→AN changes F by δb, as in F2.

4.1step 2.1step 3.1step 2.2algebra∎

For d and e choose the same alpha and compatible eta_d,eta_e; their sum solves the normalizer equation for d+e. The displayed formula then gives Fd+e=Fd+Fe term by term. For d=0 choose eta=0, giving F=0. Independence in steps 2.2 and 3.1 makes these choices irrelevant, proving additivity on cohomology. In particular if N acts trivially on A, invariance is literal equality of crossed maps, so eta=0 is allowed and F=−d(f). If N=1 or Q=1, the normalized formulas give the zero transgression.

Depends on

Used by

Dependency tree · two levels

9 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