A Spectral–Entropy Framework for the Goldbach Conjecture (v3.8)

# Summary (v3.8) This record releases a two-paper set for a programmatic entropy–spectral approach to the Goldbach problem. - **Main paper:** *Goldbach Conjecture Proof v3.8*- **Companion note:** *Goldbach Density0 v3.8* The set is organized around a strict **layering contract**: - **Proved layer (unconditional):** most-window mixing invariants (conductance / spectral gap / log–Sobolev) for a canonical **sum-chain** on dyadic windows, yielding a dyadic density–0 exceptional-set corollary.- **Conditional / roadmap layers:** optional upgrades that require clearly labeled hypotheses (HT), (EG), (H3*). No unconditional global threshold \(K_0\) is claimed. --- # Overview We define a reversible Markov chain that encodes Goldbach representations on dyadic windows and convert standard prime-distribution inputs into **mixing invariants**. The proofs are packaged as a reusable transfer pipeline: \[prime-band control conductance Φ_T gap γ_T LSI αLSI(P_T) entropy decay positivity on most windows.\] The companion note isolates a variational “density–0 closure” module under an explicitly stated energy-gap hypothesis (EG), without mixing it into the proved layer. --- # Files in this record - `Goldbach_Conjecture_Proof_v3.8.pdf`- `Goldbach_Density0_v3.8.pdf`--- # What is new in v3.8 - **Notation cross-walk and single-source-of-truth rules** are enforced (sum-chain vs pair-chain), reducing ambiguity about which kernel each theorem refers to.- The **proved vs conditional** boundary is tightened: pointwise/all-window statements are explicitly relegated to scenario modules.- The literature-positioning paragraph and table are strengthened to emphasize that the contribution is a **transfer principle / mixing-invariant packaging**, not an optimization of exceptional-set exponents. --- # Closed results (proved layer; unconditional) All unconditional results are stated for **most windows** on dyadic scales and do not require analytic continuation / functional equations. 1. **Sum-chain and mixing invariants.** Let \(S_T:=\{2n∈ 2Z: T≤ 2n≤ T+T^α\}\) and let \(Pˢᵘᵐ_T\) be the reversible Markov chain on \(S_T\) (with lazy version \( Pˢᵘᵐ_T\)). For all sufficiently large dyadic scales, for all but at most \(X/(log X)^C\) values of \(T∈[X,2X]\), one obtains polylogarithmic lower bounds of the form \[ Φ_T 1/log T, γ( Pˢᵘᵐ_T) 1/log^2 T, αLSI( Pˢᵘᵐ_T) 1/log^2 T. \] 2. **Entropy dissipation on good windows.** On those windows, the KL divergence to stationarity decays exponentially at rate \(αLSI\), giving a quantitative mixing/regularization statement. 3. **Dyadic exceptional-set corollary.** The mixing bounds imply that Goldbach positivity fails on a set of even integers of **density 0** on dyadic intervals, yielding a “most-window / dyadic density–0” Goldbach consequence. --- # Conditional / roadmap layers (explicitly isolated) These layers are included for completeness and as targets for future closure. They are **not used** in the unconditional proved layer. - **(HT) Mellin–trace / analytic continuation layer.** A heat-trace hypothesis enabling analytic continuation / functional equation statements for the Mellin–trace bridge. - **(EG) Density–0 closure energy gap (Companion).** An energy-gap hypothesis used to eliminate weakly mixed windows in a variational \(Γ\)-limit framework. - **(H3*) Window-uniform mixing (global closure scenario).** A uniform (all-window) mixing package that would imply an eventual all-even threshold \(K_0\). This record does **not** claim such a threshold unconditionally. --- # Main–Companion interface (Inputs / Outputs) - The **Main paper (proved layer)** outputs most-window mixing invariants and dyadic density–0 exceptional-set results for the **sum-chain** \(Pˢᵘᵐ_T\).- The **Companion** outputs “weakly mixed windows have log-density 0” conditional on **(EG)**.- Any upgrade from “window statements” to “integer statements” must pass through the explicitly stated bridge modules; no hidden implication is assumed. --- # Scope & non-toy status - This record should be read as a **programmatic, layer-disciplined** contribution: it supplies a rigorous transfer mechanism from prime-distribution inputs to Markov mixing invariants that are directly relevant to additive representation positivity.- The record **does not** claim an unconditional global Goldbach threshold \(K_0\).- Numerical material (if present) is **exploratory diagnostics** and is not used as a proof input. --- # How to cite Please cite as: - *Goldbach Conjecture Proof v3.8* (main paper) and- *Goldbach Density0 v3.8* (companion note),Zenodo record (this entry), version v3.8. --- # Keywords / MSC (suggested) **Keywords:** Goldbach conjecture, exceptional set, short intervals, Markov chains, conductance, spectral gap, log–Sobolev inequality, entropy dissipation, variational methods. **MSC 2020:** 11P32, 11N36, 60J10, 60J27.