Finite empirical variance from a two-coefficient comparison
Lax342547.EmpiricalVariance · concepts/Lax342547/EmpiricalVariance.lean · lax-342547
No public endorsements yet.
Loading review…
Sign in with ORCIDNatural Language Statement
Theorem
The affine-slice second-moment step keeps the independent coefficient draws and the actual bad-pair probability. This finite estimate is also valid for arbitrary bounded common tests, without polynomial assumptions on them.
Concept map
Lean source view on GitHub
| 1 | import Lax342547.FiniteSampling |
| 2 | |
| 3 | /-! |
| 4 | --- |
| 5 | title: Finite empirical variance from a two-coefficient comparison |
| 6 | type: theorem |
| 7 | --- |
| 8 | The affine-slice second-moment step keeps the independent coefficient draws |
| 9 | and the actual bad-pair probability. This finite estimate is also valid for |
| 10 | arbitrary bounded common tests, without polynomial assumptions on them. |
| 11 | -/ |
| 12 | |
| 13 | namespace Lax342547.EmpiricalVariance |
| 14 | open Lax342547.RelativeEntropy Lax342547.RetainedImages |
| 15 | open scoped BigOperators |
| 16 | |
| 17 | axiom empirical_variance {Ω A : Type} [Fintype Ω] [Fintype A] |
| 18 | (ρ : Ω → ℝ) (κ : A → ℝ) (F : Ω → A → ℝ) (good : A → A → Prop) |
| 19 | (m ε δ : ℝ) (hρ : Probability ρ) (hκ : Probability κ) |
| 20 | (hf : ∀ ω a,|F ω a| ≤ 1) (hm : |m| ≤ 1) (hε : 0 ≤ ε) |
| 21 | (hmean : ∀ a b,good a b → (∑ ω,ρ ω*F ω a) = m ∧ (∑ ω,ρ ω*F ω b) = m) |
| 22 | (hpair : ∀ a b,good a b → |(∑ ω,ρ ω*F ω a*F ω b)-m^2| ≤ ε) |
| 23 | (hbad : cellMass (fun ab : A × A => κ ab.1*κ ab.2) (fun ab => ¬ good ab.1 ab.2) ≤ δ) : |
| 24 | (∑ ω,ρ ω*((∑ a,κ a*F ω a)-m)^2) ≤ ε+4*δ |
| 25 | |
| 26 | end Lax342547.EmpiricalVariance |
| 27 |
Builds on
From Mathlib
none
Discussion
Ask a question or add context. Endorsements and structured flags are kept in the review panel above.
0 comments