Alphabeta Math
TheoremStatement: 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.

Constructive decision of the dyadic cyclotomic norm equation

Statement

For every ξ∈Z[2,1/2] there is a terminating exact algorithm deciding whether t†t=ξ has a solution t∈D=Z[ω,1/2], and returning such a solution when one exists. A supplied prime factorization of the numerator of ξ∙ξ is permissible but unnecessary for the finite-search version below. No polynomial runtime bound is asserted.

More explicitly, reject a negative real embedding; handle ξ=0 by t=0. Otherwise choose r≥0 with η=4rξ∈Z[2] and factor η into primes of O=Z[ω]. Group the prime associate classes into orbits of complex conjugation. There is a solution exactly when every conjugation-fixed class has even exponent. The conjugation-paired classes have equal exponents automatically. Allocate half of each fixed exponent, and the whole exponent to one representative of each pair, to form s. Recover a real unit v with v2=η/(s†s); then t=vs/2r is an exact solution.

Facts & Assumptions

Given: The automorphisms ∙ and † commute, and the two real embeddings are the identity and ∙ on Q(2).

[F1]

Constructive cyclotomic factorization and real unit squares proves unique factorization and a terminating factor algorithm in O, identifies O∩R=Z[2], and gives effective square roots of real units positive at both embeddings.

Proof

1.1givenalgebraconstruct

For any t∈D, both t†t=∣t∣2 and (t†t)∙=∣t∙∣2 are nonnegative. If either real embedding of ξ is negative, reject. The signs are decidable by rational arithmetic and squaring the rational coefficients of a+b2 after checking their signs. If either embedding is zero, injectivity of that embedding implies ξ=0, and the output t=0 suffices. Thus the remaining case has both embeddings strictly positive. Choose r by clearing the powers of two in the two coefficients of ξ; multiplication by the positive rational square 4r preserves their signs.

1.2F1givenalgebra

Put δ=1+ω and λ=1+2. Direct multiplication gives NO(δ)=2, δ†=ω−1δ, and δ4=2iλ2. Norm two makes δ irreducible, hence prime by [F1]. As i and λ are units, the last identity shows that 2 has just the prime associate class of δ, with multiplicity four. Every element w∈D has a unique-factorization valuation vπ(w)∈Z at each prime associate class of O, obtained from its integral numerator minus the valuations of its power-of-two denominator; only vδ(w) can be negative. Conjugation acts on these valuations by permuting the prime classes, since it is a ring automorphism. In particular, for a fixed class, vπ(w†w)=2vπ(w).

2.1step 1.1step 1.2algebra

If w†w=η∈O for w∈D, then 2vδ(w)=vδ(η)≥0, so all valuations of w are nonnegative and w∈O. To see the last implication without any assumption on units of D, write w=a/2m with a∈O; its only possible denominator class is δ, and the valuation inequality says δ4m∣a, hence 2m∣a. Consequently a solution to t†t=ξ gives an integral solution w=2rt to w†w=η. Every conjugation-fixed prime exponent of η must therefore be even. Conversely an integral solution for η gives a solution for ξ by division by 2r. This proves both necessity of the parity test and completeness of the denominator-clearing reduction.

3.1F1step 2.1algebraconstruct

Factor nonzero η by the finite algorithm of [F1]. Determine conjugation orbits by conjugating each factor and testing associates; η†=η and uniqueness of prime factorization imply that both classes in every two-element orbit occur with the same exponent. This is also an effective prime splitting test: each factor is assigned either to a fixed class or to a pair of distinct conjugate classes; no unproved classification by rational primes modulo eight is required. If a fixed exponent is odd, reject as justified in step 2.1. Otherwise, for each fixed class of exponent 2h choose its displayed prime to power h, and for each pair of exponent m choose one displayed prime to power m. Let s be the product of these choices, with empty product one. Then s†s and η have exactly the same prime-class exponents.

4.1F1step 1.1step 3.1algebraconstruct∎

Their quotient e=η/(s†s) is a unit of O by unique factorization, and is real. Its inverse is real too, so it is a unit of R=Z[2] by [F1]. Both embeddings of s†s are strictly positive, as are those of η, so both embeddings of e are positive. The unit-square construction of [F1] therefore returns v∈R with v2=e. Because v†=v, (vs)†(vs)=es†s=η, and division by 2r gives the claimed t. All loops terminate: denominator clearing is finite, factorization terminates, associate tests are exact, the factor list is finite, and square-root recovery terminates for a positive real unit. Together with the rejection cases and necessity this proves the entire decision and construction claim.

Remarks

The proof supplies a complete terminating mode. Ross–Selinger Appendix C uses rational-prime splitting and modular square-root algorithms for a faster mode. Its expected polynomial runtime is a separate complexity assertion; it does not follow from the finite-search construction here.

Depends on

Used by

Dependency tree · two levels

5 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