Main theorem: the Kalton-Roberts upper bound #
The final proof that the Kalton-Roberts constant is less than 9919 / 500.
Case 1 details #
Case 1 of the proof of Theorem 1.1 (q ≤ q₀).
Given M, D' satisfying the two recombination inequalities from E₁ and E₂:
M ≤ A₁ − (1/2) D' (Equation (12), from one-sided recombination with E₁)
M ≤ 15 + (7/3) D' (Equation (13), from one-sided recombination with E₂)
we conclude M ≤ C₁.
Reference: Case 1 in Section 5 of the companion paper, Equations (12)–(14).
Case 2 details #
Case 2 of the proof of Theorem 1.1 (q ≥ q₀).
Given M, Y satisfying the two recombination inequalities from E₃ and E₄:
M ≤ A₂ − (2/5) Y (Equation (15), from two-sided recombination with E₃)
M ≤ 47/3 + (11/6) Y (Equation (16), from two-sided recombination with E₄)
we conclude M ≤ C₂.
Reference: Case 2 in Section 5 of the companion paper, Equations (15)–(17).
The u-bound (Equation (3)) #
The finite normalised theorem #
Core distance bound: distToAdditive f ≤ C₂.
The proof depends on the full pipeline:
- Dual certificate (proved in
DualCert.lean) - Weighted collection construction (proved in
Collections.lean) - Mixed intersection construction (Corollary 3.1)
- Expander recombination (Lemmas 3.2–3.3)
- Expander existence via Pippenger entropy (Lemma 4.1)
- Case analysis and balancing (Section 5)
The theorem distToAdditive_le_C₂_from_pipeline in Pipeline.lean
makes explicit each missing ingredient as a hypothesis. Here we
instantiate those hypotheses from the concrete theorems.
The normalized theorem on arbitrary Boolean algebras, obtained from the finite powerset theorem by the compact reduction.
Theorem 1.1 in its scaled Boolean-algebra form.
The paper's set-algebra formulation. A set algebra is represented as a
Boolean subalgebra of Set Ω.
Main theorem #
Theorem 1.1 (Main theorem). The Kalton–Roberts constant satisfies
K_KR ≤ C₂ = 694198146664396294486127753 / 34994834677886019996000000.
Consequently, K_KR < 9919/500 = 19.838.
Reference: Theorem 1.1 in Section 1 of the companion paper.
Theorem 1.1, headline inequality. Reference: second display of Theorem 1.1 in the companion paper.