AFTD Auto-Formalizing Theoretical Domains

hasBernsteinMGF_centered_chi_squared

theoremverifiedimported Lean

Bernstein MGF bound for a centered chi-squared summand. Fix an integer $k \ge 1$, and let $Y$ be a real-valued random variable on a probability space $(\Omega, \mu)$ whose law is the centered Gaussian $\mathcal{N}(0, 1/k)$ of mean $0$ and variance $1/k$. Then the centered square $Y^2 - 1/k$ satisfies the Bernstein MGF condition on $(\Omega, \mu)$ with quadratic-growth parameter $c = 2/k^2$ and radius $t_{\max} = k/4$.

Statement

set_option relaxedAutoImplicit false in
set_option autoImplicit false in
open MeasureTheory ProbabilityTheory Real NNReal Matrix Finset in
variable {d k : ℕ} in
theorem hasBernsteinMGF_centered_chi_squared
    {Ω : Type*} [MeasurableSpace Ω] (μ : Measure Ω) [IsProbabilityMeasure μ]
    (k : ℕ) (hk : 0 < k)
    (Y : Ω → ℝ) (hY_meas : Measurable Y)
    (hY_law : Measure.map Y μ = gaussianReal 0 ⟨1 / k, by positivity⟩) :
    HasBernsteinMGF (fun ω => (Y ω) ^ 2 - 1 / k) μ (2 / (k : ℝ) ^ 2) ((k : ℝ) / 4)
Source
TCSlib, `hasBernsteinMGF_centered_chi_squared`
Lean proof
By Ganesh Sankar, from github.com/Shilun-Allan-Li/tcslib at c8a02591b1e5 (Copyright (c) 2026 Ganesh Sankar. All rights reserved, Apache-2.0). Compiled here against leanprover/lean4:v4.33.1, Mathlib v4.33.1; written for leanprover/lean4:v4.25.0.
Verified
07 Oct 2026
Axioms
Classical.choiceQuot.soundpropext
Built on 2
  • centered_chi_squared_stepCentered chi-squared MGF bound. Let $(\Omega,\mu)$ be a probability space and let $k$ be a positive natural number. Suppose…
  • ProbabilityTheory.HasBernsteinMGF$\texttt{ProbabilityTheory.HasBernsteinMGF}\ X\ \mu\ c\ t_{\max}$ is a predicate on a random variable $X : \Omega \to \mathbb{R}$, a…
Used by 1
  • chi_squared_tailChi-squared tail bound for independent $N(0,1/k)$ variables. Let $\mu$ be a probability measure on $\Omega$, let $k \ge 1$, and let…

Lean source view module on GitHub

set_option relaxedAutoImplicit false inset_option autoImplicit false inopen MeasureTheory ProbabilityTheory Real NNReal Matrix Finset invariable {d k : ℕ} in/-- **Bernstein MGF instance for the centered chi-squared summand.** For a measurable real random variable `Y` on a probability space with law `N(0, 1/k)` (`k > 0`), the centered square `Y² − 1/k` has a Bernstein-type MGF with parameters `(2/k², k/4)`: `exp(t(Y² − 1/k))` is integrable and its mean is at most `exp((2/k²) t²)` for `|t| ≤ k/4`. This is the sub-exponential property of a centered chi-squared summand, [Ver18, Lemma 2.7.6] (Gaussian case) with the explicit constants from [DG03, proof of Lemma 2.2]; it packages `centered_chi_squared_step` into the abstract Bernstein form so that the sum and tail-bound machinery in `TCSlib.LearningTheory.JohnsonLindenstrauss.Bernstein` applies uniformly to Gaussian, Rademacher, and other sub-Gaussian families. -/theorem hasBernsteinMGF_centered_chi_squared    {Ω : Type*} [MeasurableSpace Ω] (μ : Measure Ω) [IsProbabilityMeasure μ]    (k : ℕ) (hk : 0 < k)    (Y : Ω → ℝ) (hY_meas : Measurable Y)    (hY_law : Measure.map Y μ = gaussianReal 0 ⟨1 / k, by positivity⟩) :    HasBernsteinMGF (fun ω => (Y ω) ^ 2 - 1 / k) μ (2 / (k : ℝ) ^ 2) ((k : ℝ) / 4) := by  refine ⟨?_, ?_⟩  · intro t ht    exact (centered_chi_squared_step μ k hk Y hY_meas hY_law t ht).1  · intro t ht    have h := (centered_chi_squared_step μ k hk Y hY_meas hY_law t ht).2    -- `mgf ≤ exp(2 t²/k²)`. Match `c = 2/k²`, so `c · t² = 2 t²/k²`. ✓    convert h using 2    ring
See it in the graphDiscuss this resultChallenge it