Zero Poisson count gives a deterministic observable
ProvedKSATSingleFlip.Connect.poisson_zero_count_deterministicIf the count has Poisson mean zero, and the observable is constant on the zero-count event, then its probability of falling any positive amount below its mean is zero. This supplies the zero-mean case in Lemma 2.6.
Preamble
import Mathlib set_option autoImplicit false open MeasureTheory ProbabilityTheory Set open scoped NNReal ENNReal
Formal statement
theorem KSATSingleFlip.Connect.poisson_zero_count_deterministic {Ω : Type} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P]
(N : Ω → ℕ) (hN : Measurable N) (hNlaw : Measure.map N P = poissonMeasure 0)
(Y : Ω → ℝ) (c u : ℝ) (hu : 0 < u) (hzero : ∀ ω, N ω = 0 → Y ω = c) :
P.real {ω | Y ω ≤ (∫ ω, Y ω ∂P)-u} = 0 := by sorrySource
Wang–Yin–Yu–Zhang, Single-Variable Connectivity in Random k-SAT, Lemma 2.6, https://github.com/sigact-stoc/stoc2027/releases/download/papers-1/737.pdf