Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedjudge pass (gpt-6.1-sol)
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.

The Ocneanu Markov trace exists and is unique

Statement

Let H(1)⊂H(2)⊂⋯ be the type-A Hecke tower over Λ=Z[v±1,z] of The Markov trace on the type-A Hecke tower. Then there exists a unique Markov trace (tr⁡n)n≥1 on this tower in the sense of The Markov trace on the type-A Hecke tower. Moreover it satisfies, for all n≥1, all x,y∈H(n) and all 1≤i≤n−1:

  • (a) tr⁡n(Ti)=z;
  • (b) tr⁡n+1(xTny)=z tr⁡n(xy);
  • (c) tr⁡n is determined by (M1)--(M4) alone; it takes values in Λ and is computed by iterating (b) along the free basis of The Hecke tower is free over the previous level.

Facts & Assumptions

Given: The Hecke tower H(1)⊂H(2)⊂⋯ over Λ=Z[v±1,z]. No choice principle is used.

[F1]

Conditions (M1)--(M4) of a Markov trace and the equivalence of the two forms of (M4) (The Markov trace on the type-A Hecke tower).

[F2]

For every n≥1, H(n+1)=⨁i=0nH(n)Tw(i), where w(i)=snsn−1⋯sn−i+1 and Tw(i)=TnTn−1⋯Tn−i+1 for i≥1; each element has a unique expression ∑ixiTw(i) with xi∈H(n). Moreover H(n+1)=H(n)⊕(H(n)⊗H(n−1)H(n)) as H(n)-bimodules, so every element of H(n+1) determines a unique pair (a,ξ) with a∈H(n) and ξ∈H(n)⊗H(n−1)H(n); a finite sum representing ξ is taken modulo the tensor relations, including xTnhy=xhTny for h∈H(n−1) (The Hecke tower is free over the previous level).

[F3]

H(n) has Λ-basis {Tw:w∈Sn}, and for w∈Sn, TwTi=Twsi or (v−1)Tw+vTwsi according as ℓ(wsi)=ℓ(w)+1 or ℓ(wsi)=ℓ(w)−1; the quadratic relation is Ti2=(v−1)Ti+v (The standard basis of the generic type-A Hecke algebra, The generic type-A Hecke algebra).

Proof

1.1F1F2algebra

Uniqueness. Suppose (tr⁡n) is a Markov trace. The base is H(1)=Λ, where tr⁡1(λ)=λ by (M1) and Λ-linearity. For n≥2, [F2] at level n−1 gives each y∈H(n) the unique expansion y=∑i=0n−1xiTw(i) with xi∈H(n−1) and Tw(i)=Tn−1⋯Tn−i. By (M2), tr⁡n(x0)=tr⁡n−1(x0); by (M3) and the two-sided form of (M4), tr⁡n(xiTw(i))=tr⁡n(Tw(i)xi)=ztr⁡n−1(Tn−2⋯Tn−ixi) for i≥1, an element of H(n−1) on which tr⁡n−1 is already defined. Hence tr⁡n is determined by tr⁡n−1; induction gives uniqueness and (c).

1.2F1F2construct

Recursive construction. Define tr⁡1:Λ→Λ by tr⁡1(a)=a. Suppose tr⁡n is defined. The bimodule isomorphism in [F2] is induced by μn(x⊗y)=xTny and gives H(n+1)=H(n)⊕im⁡(μn). The Λ-linear map τn:H(n)⊗H(n−1)H(n)→Λ, x⊗y↦tr⁡n(xy), is balanced: (xh)⊗y=x⊗(hy) for h∈H(n−1), and both tensors map to tr⁡n(xhy). Define tr⁡n+1(a+μn(ξ)):=tr⁡n(a)+zτn(ξ),a∈H(n), ξ∈H(n)⊗H(n−1)H(n). The direct-sum decomposition and the isomorphism μn make this definition well defined and Λ-linear. Restriction to H(n) gives (M2), and tr⁡n+1(1)=tr⁡n(1) gives (M1). For each i<n, repeated restriction gives tr⁡n(Ti)=tr⁡i+1(Ti)=z by the recursion at level i, proving (a). By construction, tr⁡n+1(xTny)=ztr⁡n(xy)(x,y∈H(n)); this is the two-sided recursion, and y=1 gives (M4). Iterating it along the left basis of [F2] gives the recursive formula in (c).

2.1F1F2step 1.2

Cyclicity: reduction. We prove cyclicity by induction. The base H(1)=Λ is commutative, and H(2) is generated over Λ by the single element T1, so its trace is cyclic. For n≥2, assume tr⁡n is cyclic and consider H(n+1)=H(n)⊕In, where In is the H(n)-sub-bimodule spanned by xTny, as in [F2]. If a,b∈H(n), cyclicity is the induction hypothesis. If a∈H(n) and b=xTny∈In, then the construction in step 1.2 gives tr⁡n+1(ab)=ztr⁡n(axy) and tr⁡n+1(ba)=ztr⁡n(xya), equal by induction; linearity handles sums in In. Thus it remains the case a=xTny, b=uTnv with x,y,u,v∈H(n). Applying the already proved one-in-H(n) case to the outer factors reduces tr⁡n+1(ab)=tr⁡n+1(TnXTnY) and tr⁡n+1(ba)=tr⁡n+1(TnYTnX), where X=yu and Y=vx. By that same case, this is equivalent to tr⁡n+1(TnXTnY)=tr⁡n+1(XTnYTn)(X,Y∈H(n)). It remains to prove this identity.

3.1F1F2F3step 1.2step 2.1algebra∎

The final cases. Use [F2] at level n−1 to write H(n)=H(n−1)⊕H(n−1)Tn−1H(n−1); the balance here is over H(n−2), since Tn−1 commutes with H(n−2). If X,Y∈H(n−1), then Tn commutes with both and the desired identity follows from the quadratic relation for Tn. For X=x′Tn−1x′′ with x′,x′′∈H(n−1) and Y∈H(n−1), commuting Tn past x′,x′′,Y and applying the braid relation gives tr⁡n+1(TnXTnY)=ztr⁡n(x′Tn−12x′′Y). On the other side, commute Tn past x′′ and Y, expand Tn2, and use the two-sided recursion from step 1.2 to obtain tr⁡n+1(XTnYTn)=(v−1)ztr⁡n(x′Tn−1x′′Y)+vtr⁡n(x′Tn−1x′′Y). The level-n recursion gives tr⁡n(x′Tn−1x′′Y)=ztr⁡n−1(x′x′′Y), while restriction gives tr⁡n(x′x′′Y)=tr⁡n−1(x′x′′Y). Expanding Tn−12 in the first display therefore yields the same expression as the right side. If X∈H(n−1) and Y∈H(n−1)Tn−1H(n−1), write Y=y′Tn−1y′′ and put a:=Xy′∈H(n−1). Since Tn commutes with X,y′,y′′, the quadratic relation and two-sided recursion give tr⁡n+1(TnXTnY)=(v−1)ztr⁡n(aTn−1y′′)+vtr⁡n(aTn−1y′′),tr⁡n+1(XTnYTn)=z(v−1)tr⁡n(aTn−1y′′)+zvtr⁡n(ay′′). By (M2) and the two-sided recursion at level n, tr⁡n(ay′′)=tr⁡n−1(ay′′) and tr⁡n(aTn−1y′′)=ztr⁡n−1(ay′′); hence these expressions agree. Finally let X=x′Tn−1x′′ and Y=y′Tn−1y′′ with all four coefficients in H(n−1). Braid, commutation, and the two-sided recursion give tr⁡n+1(TnXTnY)=ztr⁡n(x′Tn−12x′′y′Tn−1y′′),tr⁡n+1(XTnYTn)=ztr⁡n(x′Tn−1x′′y′Tn−12y′′). After expanding the squared generators, the terms with coefficient z(v−1) agree. The remaining terms agree because the level-n recursion and the induction hypotheses that tr⁡n and tr⁡n−1 are cyclic give tr⁡n(x′x′′y′Tn−1y′′)=ztr⁡n−1(x′x′′y′y′′),tr⁡n(x′Tn−1x′′y′y′′)=ztr⁡n−1(x′x′′y′y′′). Thus the central identity holds in every case, (M3) follows, and the induction is complete.

Depends on

Used by

Dependency tree · two levels

17 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