trureturingGitHub MATHEMATICAL ATLAS

RESEARCH / Focused target

Parity of the Catalan-composition series A374570

Literature status: not recheckedRepository record: proved

RELEASED FOUNDATIONS / PROPOSED CONNECTION

Research connections

1 released anchors
All source anchors (1)

Problem

The question

OEIS A374570 (Paul D. Hanna, Jul 11 2024), quoted verbatim from Library/Arith/hanna2024a374570.md:

NAME:

Expansion of g.f. A(x) satisfying A(x)^2 = A( A(x)*C(x) ), where C(x) = x + C(x)^2 is the Catalan function (A000108).

COMMENT:

Conjecture: for n > 1, a(n) is odd iff n = 2^k + 1 for k >= 0.

The normalization is A(0) = 0, a(1) = 1, and C(0) = 0.

Motivation

Our foothold

This is a first-tier recent OEIS conjecture. The KPI is open problems resolved, not modules, seats, or finite checks. The target is the assertion for every natural index n > 1.

Gap

Missing bridges

The supplied search reports no proof found: the search seat read the OEIS entry and revision history on 2026-09-08 and searched the identifier on arXiv, MathOverflow, and GitHub. This Stage-B seat had no network and did not repeat those searches. This bounded search is not an exhaustive literature census.

Route

Proposed approach

Set C = X * catalanUnit, using Mathlib's PowerSeries.Catalan; catalan_equation and catalan_unique identify the normalized Catalan function. Construct the integer solution A = X * B by degree-contracting stabilization of B -> catalanUnit * B(X^2 * B * catalanUnit). The results generating_equation and generating_unique establish existence and uniqueness of the normalized solution to the exact functional equation.

Over ZMod 2, the live cancellation chain is A(X^2) = A(A*C) implies A*C = X^2 implies A = X*(1+C). Outer cancellation uses Mathlib's compositional-inverse API. The Catalan identity gives C*(1+C) = X; writing C as X times a unit justifies the remaining cancellation. Strong induction on the characteristic-two coefficient contraction gives binary_catalan: the support of C consists exactly of powers of two. Hence, for n > 1, a(n) is odd if and only if n - 1 = 2^k, equivalently n = 2^k + 1, as stated by hanna_conjecture. The module has no D5 import and has generality G.

Falsifier

What would falsify this route

A counterexample index n > 1 at which Odd (a n) disagrees with exists k : Nat, n = 2^k + 1 would contradict the assertion. The orchestrator's exact numerical check is supporting evidence only; no finite bound proves this universally quantified statement.

Evidence

Evidence to collect

  • Lean module: D5/S1/Recurrence/Invariants/CatalanCompositionSquareParity.lean.
  • Main theorem: hanna_conjecture.
  • Companions: catalan_equation, catalan_unique, generating_equation, generating_unique, and binary_catalan.
  • Axioms: std3 (propext, Classical.choice, Quot.sound), as reported for each public theorem by the implementation seat.

Triage

Scope assessment

theorem. The formal proof closes the universal assertion recorded by OEIS for the normalized integer generating series.

ASSUMED-UNVERIFIED

Unverified assumptions

The OEIS quotations and attribution were supplied by the orchestrator and copied from the Library note. The OEIS revision history was read by the search seat, not this seat. The literature scope was the OEIS entry and history plus identifier searches on arXiv, MathOverflow, and GitHub, as reported above; this seat had no network. First-publication priority and the source-to-Lean identification are not kernel-checked facts. The std3 report and numerical check were supplied evidence, not checks rerun by Stage B.

Dossier & anchor history

Loading release observations...