Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Gaussian samples: ∥C−1/2(μ^−μ)∥≤χ/N\|C^{-1/2}(\hat\mu-\mu)\|\le\chi/\sqrt N∥C−1/2(μ^​−μ)∥≤χ/N​ and (1−ζ)C⪯C^⪯(1+ζ)C(1-\zeta)C\preceq\hat C\preceq(1+\zeta)C(1−ζ)C⪯C^⪯(1+ζ)C with probability ≥1−3e−t2/2\ge1-3e^{-t^2/2}≥1−3e−t2/2

Proved
BosonicLearn.gaussian_sample_concentration

by Senrui Chen · Nov 13, 2026 · Mathlib 0df444a (Lean v4.33.1)

bosonic-gaussian-statesquantum-informationquantum-learning

Mean and covariance of NNN Gaussian samples concentrate in relative (whitened) form, with failure probability 3e−t2/23e^{-t^2/2}3e−t2/2.

Let CCC be a positive definite 2n×2n2n\times2n2n×2n real matrix, μ∈R2n\mu\in\mathbb R^{2n}μ∈R2n, N≥1N\ge1N≥1 and t≥0t\ge0t≥0. Put

χ=2n+t,ζ=2χN+2χ2N.\chi=\sqrt{2n}+t,\qquad \zeta=\frac{2\chi}{\sqrt N}+\frac{2\chi^2}{N}.χ=2n​+t,ζ=N​2χ​+N2χ2​.

Let Y1,…,YNY_1,\dots,Y_NY1​,…,YN​ be i.i.d. N(μ,C)\mathcal N(\mu,C)N(μ,C), μ^=1N∑jYj\hat\mu=\frac1N\sum_jY_jμ^​=N1​∑j​Yj​ and C^=1N∑j(Yj−μ^)(Yj−μ^)⊤\hat C=\frac1N\sum_j(Y_j-\hat\mu)(Y_j-\hat\mu)^\topC^=N1​∑j​(Yj​−μ^​)(Yj​−μ^​)⊤. Then with probability at least 1−3e−t2/21-3e^{-t^2/2}1−3e−t2/2, simultaneously

(μ^−μ)⊤C−1(μ^−μ)≤χ2Nand(1−ζ) C⪯C^⪯(1+ζ) C.(\hat\mu-\mu)^\top C^{-1}(\hat\mu-\mu)\le\frac{\chi^2}{N}\qquad\text{and}\qquad(1-\zeta)\,C\preceq\hat C\preceq(1+\zeta)\,C .(μ^​−μ)⊤C−1(μ^​−μ)≤Nχ2​and(1−ζ)C⪯C^⪯(1+ζ)C.

Relation to Eq. (210). This is the event of Eq. (210) (BosonicLearn.eq_210) with the deviation parameter ttt left free. The three terms of the failure probability come from the norm of the whitened sample mean and from the smallest and largest singular values of the whitened N×2nN\times2nN×2n data matrix. Taking t=2log⁡(3/δ)t=\sqrt{2\log(3/\delta)}t=2log(3/δ)​ gives probability at least 1−δ1-\delta1−δ; taking t=2log⁡(2/δ)t=\sqrt{2\log(2/\delta)}t=2log(2/δ)​, the choice of Eq. (209), this argument gives 1−32δ1-\tfrac32\delta1−23​δ. For the proof of Theorem 8 the difference only changes the universal constant.

Role. It replaces BosonicLearn.eq_210 in the proof of Theorem 8 (heterodyne tomography), where C=Σ+121C=\Sigma+\tfrac12\mathbf 1C=Σ+21​1 is the covariance of the heterodyne outcomes.

Formalization Note Same event set as eq_210: samples are Measure.pi fun _ : Fin N => multivariateGaussian (WithLp.toLp 2 μ) C on EuclideanSpace ℝ (Fin n ⊕ Fin n), Loewner order is Matrix.PosSemidef of the difference, and the first inequality is the squared form of ∥C−1/2(μ^−μ)∥2≤χ/N\|C^{-1/2}(\hat\mu-\mu)\|_2\le\chi/\sqrt N∥C−1/2(μ^​−μ)∥2​≤χ/N​. The bound is 1 - ENNReal.ofReal (3 * exp (-t^2/2)). The proof imports the proved theorems GaussianMatrix.extreme_singular_values_deviation, GaussianMatrix.gaussian_concentration and GaussianMatrix.frobenius_second_moment of the mission series on Gaussian random matrices.

Preamble
import Mathlib

open MeasureTheory ProbabilityTheory
open scoped ENNReal Matrix
Formal statement
namespace BosonicLearn

/-- **Mean and covariance concentration for `N` Gaussian samples, with failure probability `3e^{−t²/2}`.**
Chen, Anna Mele, Anna Mele, Preskill, *When are bosonic Gaussian states classical to learn?*, arXiv:2609.26705, §7.2, proof of Theorem 8, Eqs. (208)–(210), p. 61: "Set χ := √(2n) + √(2 log(2/δ)), ζ := 2χ/√N + 2χ²/N. The
Gaussian concentration bound of [BMEM25, Lemma 9, Eqs. (91), (94), and (95)] implies that, with probability at
least 1 − δ, ‖C^{−1/2}(µ̂ − µ)‖_2 ≤ χ/√N, (1 − ζ)C ⪯ Ĉ ⪯ (1 + ζ)C."

Let `C` be a positive definite `2n × 2n` matrix, `µ ∈ ℝ²ⁿ`, `N ≥ 1`, `t ≥ 0`, `χ = √(2n) + t`,
`ζ = 2χ/√N + 2χ²/N`, and `Y₁, …, Y_N` i.i.d. `N(µ, C)` with empirical mean `µ̂` and centered empirical
covariance `Ĉ`. Then with probability at least `1 − 3e^{−t²/2}` both
`(µ̂ − µ)ᵀ C⁻¹ (µ̂ − µ) ≤ χ²/N` and `(1 − ζ)C ⪯ Ĉ ⪯ (1 + ζ)C` hold.

**Formalization Note.** Same event as `eq_210`, with the deviation parameter `t` free and the failure
probability `3e^{−t²/2}`: one term for the norm of the whitened sample mean and two for the extreme singular
values of the whitened data matrix. Taking `t = √(2 log(3/δ))` gives probability at least `1 − δ`; with
`t = √(2 log(2/δ))` as in Eq. (209) this argument gives `1 − 3δ/2`. -/
theorem gaussian_sample_concentration {n : ℕ} (N : ℕ) (hN : 1 ≤ N) (t : ℝ) (ht : 0 ≤ t)
    (μ : Fin n ⊕ Fin n → ℝ) (C : Matrix (Fin n ⊕ Fin n) (Fin n ⊕ Fin n) ℝ) (hC : C.PosDef) :
    let χ : ℝ := Real.sqrt (2 * n) + t
    let ζ : ℝ := 2 * χ / Real.sqrt N + 2 * χ ^ 2 / N
    1 - ENNReal.ofReal (3 * Real.exp (-t ^ 2 / 2)) ≤
      (Measure.pi fun _ : Fin N => multivariateGaussian (WithLp.toLp 2 μ) C)
        {Y | let μhat : Fin n ⊕ Fin n → ℝ := fun a => (1 / (N : ℝ)) * ∑ j, WithLp.ofLp (Y j) a
              let Chat : Matrix (Fin n ⊕ Fin n) (Fin n ⊕ Fin n) ℝ := Matrix.of fun a b =>
                (1 / (N : ℝ)) * ∑ j, (WithLp.ofLp (Y j) a - μhat a) * (WithLp.ofLp (Y j) b - μhat b)
              (μhat - μ) ⬝ᵥ (C⁻¹ *ᵥ (μhat - μ)) ≤ χ ^ 2 / N ∧
                (Chat - (1 - ζ) • C).PosSemidef ∧ ((1 + ζ) • C - Chat).PosSemidef} := by sorry

end BosonicLearn
Source
Chen, Anna Mele, Anna Mele, Preskill, When are bosonic Gaussian states classical to learn?, arXiv:2609.26705, §7.2, proof of Theorem 8, Eqs. (208)–(210), p. 61 (mean and covariance concentration, [BMEM25, Lemma 9])

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me