c_G
For a real number ρ, the Gaussian majority limit stability constant c_G(ρ) : ℝ is defined as 2 * ((∫ g, Real.sqrt (1 - (real_erf (ρ * g / Real.sqrt (2 * (1 - ρ ^ 2)))) ^ 2) ∂(ProbabilityTheory.gaussianReal 0 1)) - Real.sqrt (1 - ρ ^ 2)).
Statement
open MeasureTheory ProbabilityTheory in noncomputable def c_G (ρ : ℝ) : ℝ
- Defined in
- OP-797
- Verified
- 01 Nov 2026
- Axioms
- none listed
- Built on 1
- real_erfThe Gauss error function erf(z) is defined as (2 / sqrt(pi)) times the integral from 0 to z of exp(-t^2).
Read back from the Lean
c_G is a function from ℝ to ℝ defined for any $\rho \in \mathbb{R}$ by:
$$c_G(\rho) = 2 \left( \left( \int g, \sqrt{1 - \left(\operatorname{real\_erf}\left(\frac{\rho \cdot g}{\sqrt{2(1 - \rho^2)}}\right)\right)^2} \, \partial(\operatorname{gaussianReal}(0, 1)) \right) - \sqrt{1 - \rho^2} \right)$$
where the integration variable $g : \mathbb{R}$ is integrated with respect to the standard Gaussian measure on $\mathbb{R}$ with mean $0$ and variance $1$.
Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.
Lean source view module on GitHub
open MeasureTheory ProbabilityTheory in /-- The Gaussian majority limit stability constant c_G(rho). -/ noncomputable def c_G (ρ : ℝ) : ℝ := 2 * ((∫ g, Real.sqrt (1 - (real_erf (ρ * g / Real.sqrt (2 * (1 - ρ ^ 2)))) ^ 2) ∂(ProbabilityTheory.gaussianReal 0 1)) - Real.sqrt (1 - ρ ^ 2))