Small normalized likelihood mean from a compact variational envelope
ProvedZeroFreeRRG.Hardcore.overlap_likelihood_mean_smallLet be compact and let continuous have on , unique zero , and . Suppose that for each and , the sequence is eventually bounded above by . Then for every some satisfies for all sufficiently large . This is the abstract passage from the variational upper bound (29) to the small expected reveal log-likelihood used in Lemma 5.6.
Preamble
import Mathlib set_option autoImplicit false open Set
Formal statement
theorem ZeroFreeRRG.Hardcore.overlap_likelihood_mean_small {α : Type} [TopologicalSpace α] (K : Set α) (hK : IsCompact K)
(f g : α → ℝ) (hf : Continuous f) (hg : Continuous g) (x₀ : α) (hx₀ : x₀ ∈ K)
(hfnonpos : ∀ x ∈ K, f x ≤ 0) (hfzero : f x₀ = 0)
(hunique : ∀ x ∈ K, f x = 0 → x = x₀) (hgzero : g x₀ = 0)
(E : ℝ → ℕ → ℝ)
(hbound : ∀ s : ℝ, 0 < s → s < 1 → ∀ δ : ℝ, 0 < δ →
∃ N : ℕ, ∀ n ≥ N, ∃ x ∈ K, E s n ≤ f x + s*g x + δ) :
∀ ε : ℝ, 0 < ε → ∃ s : ℝ, 0 < s ∧ s < 1 ∧
∃ N : ℕ, ∀ n ≥ N, E s n ≤ ε*s := by sorrySource
Auxiliary step in Chen–Wang–Zhang, Zero-Freeness for Random Regular Graphs up to the Replica Symmetry Breaking Threshold, https://github.com/sigact-stoc/stoc2027/releases/download/papers-1/1461.pdf, p. 39, Lemma 5.6 and equation (29).