Back to search
2006.11634

A Fractional 3n+1 Conjecture

Éric Brier, Rémi Géraud-Stewart, David Naccache

wronghigh confidence
Category
math.DS
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:55 AM

Audit review

The paper states and claims to prove that for every real u0 in [0,100], the Δ-trajectory either converges to 0 or eventually loops on a length‑29 cycle (Theorem 1) . Its proof shows that on an interval I one has Δ^29(x) = (3^17 x + B)/2^29, hence Δ^29 is a contraction with a unique fixed point x0, and then asserts that any seed eventually “enters the cycle” after mapping into I . This conflates asymptotic attraction (lim i→∞ Δ^{29 i}(x) = x0) with eventual periodicity, which is false for irrational seeds. Indeed, Δ maps irrationals to irrationals, and any periodic point of this piecewise affine map must be rational; hence positive irrational seeds can never be eventually periodic. The paper quantifies over all reals in [0,100], so its theorem as stated is wrong. The candidate solution correctly identifies this logical flaw and the rationality/irrationality barrier, and also (independently) confirms the existence of the 29‑cycle (though its reported numerator differs from the paper’s; the paper’s x0 = 616136875/(2^29−3^17) is consistent with its own computation) .

Referee report (LaTeX)

\textbf{Recommendation:} major revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The main claim is overstated: the provided argument establishes attraction to a 29-cycle for seeds that enter a certain interval, but not eventual periodicity for every real seed in the stated range. As irrational seeds cannot be periodic under Δ, the theorem as stated is false. With a corrected statement (either describing ω-limit sets or restricting to rational seeds) the work would be a solid contribution documenting an interesting dynamical system and computational method.