Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Equation 2.2.10 — Variance statistic under dilation

Proved
TroppMatrixConcentration.dilation_variance

by tc · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

matrix-concentrationrandom-matrices

Let ZZZ be a measurable complex m×nm\times nm×n random matrix with finite second moment on a probability space, where m,n≥1m,n\ge1m,n≥1. Put C=Z−EZC=Z-\mathbb EZC=Z−EZ. Then

∥EH(C)2∥=max⁡{∥ECC∗∥,∥EC∗C∥}.\left\|\mathbb E H(C)^2\right\|=\max\{\|\mathbb E CC^*\|,\|\mathbb E C^*C\|\}.​EH(C)2​=max{∥ECC∗∥,∥EC∗C∥}.

This is the equality of the matrix variance statistic of a Hermitian dilation and that of the original rectangular random matrix. Centering is explicit and no zero-mean assumption is needed.

Preamble
import Definitions.Def_TroppMatrixConcentration_probability
import Definitions.Def_TroppMatrixConcentration_dilation
import Mathlib.Analysis.Convex.Function

open MeasureTheory ProbabilityTheory
open scoped Matrix.Norms.L2Operator

Formal statement
namespace TroppMatrixConcentration

theorem dilation_variance {Ω : Type*} [MeasurableSpace Ω]
    (μ : Measure Ω) [IsProbabilityMeasure μ] {m n : ℕ} [NeZero m] [NeZero n]
    (Z : Ω → Matrix (Fin m) (Fin n) ℂ) (hMeas : Measurable Z)
    (hL2 : MemLp Z 2 μ) :
    hermitianSecondMoment μ (fun ω => dilation (Z ω - ∫ ω', Z ω' ∂μ)) =
      rectSecondMoment μ (fun ω => Z ω - ∫ ω', Z ω' ∂μ) := by sorry

end TroppMatrixConcentration
Source
Joel A. Tropp, An Introduction to Matrix Concentration Inequalities, arXiv:1501.01571v1 (7 January 2015); https://arxiv.org/abs/1501.01571v1; Section 2.2.8, equations (2.2.7–10), printed pp. 28–29.
Read-back

What the Lean code literally says, in plain math · gpt-6

Let Ω\OmegaΩ be any measurable space, let μ\muμ be a probability measure on it, let m,nm,nm,n be positive natural numbers, and let Z:Ω→Cm×nZ:\Omega\to\mathbb C^{m\times n}Z:Ω→Cm×n be measurable and square integrable, meaning that ∫Ω∥Z(ω)∥op2 dμ(ω)<∞\int_\Omega\|Z(\omega)\|_{\mathrm{op}}^2\,d\mu(\omega)<\infty∫Ω​∥Z(ω)∥op2​dμ(ω)<∞, where ∥⋅∥op\|\cdot\|_{\mathrm{op}}∥⋅∥op​ denotes the operator norm for the Euclidean norms on complex coordinate spaces. Write M=∫ΩZ(ω) dμ(ω)M=\int_\Omega Z(\omega)\,d\mu(\omega)M=∫Ω​Z(ω)dμ(ω) for the Bochner mean, set A(ω)=Z(ω)−MA(\omega)=Z(\omega)-MA(ω)=Z(ω)−M, and define the dilation of an m×nm\times nm×n matrix BBB to be the (m+n)×(m+n)(m+n)\times(m+n)(m+n)×(m+n) block matrix D(B)=(0BB∗0)D(B)=\begin{pmatrix}0&B\\B^*&0\end{pmatrix}D(B)=(0B∗​B0​), indexed by the disjoint union of the row and column index sets, where B∗B^*B∗ is the conjugate transpose. Then ∥∫ΩD(A(ω))2 dμ(ω)∥op=max⁡ ⁣{∥∫ΩA(ω)A(ω)∗ dμ(ω)∥op,∥∫ΩA(ω)∗A(ω) dμ(ω)∥op}\left\|\int_\Omega D(A(\omega))^2\,d\mu(\omega)\right\|_{\mathrm{op}}=\max\!\left\{\left\|\int_\Omega A(\omega)A(\omega)^*\,d\mu(\omega)\right\|_{\mathrm{op}},\left\|\int_\Omega A(\omega)^*A(\omega)\,d\mu(\omega)\right\|_{\mathrm{op}}\right\}​∫Ω​D(A(ω))2dμ(ω)​op​=max{​∫Ω​A(ω)A(ω)∗dμ(ω)​op​,​∫Ω​A(ω)∗A(ω)dμ(ω)​op​}. Here the left side is the defined Hermitian second moment of the dilated centered matrix, and the right side is the defined rectangular second moment of the centered matrix; every integral is a Bochner integral and the square on the dilation is matrix multiplication. The quantifiers exclude zero row or column dimension, and the probability-measure hypothesis excludes an empty underlying space; they include constant or almost everywhere constant ZZZ, for which both sides are zero, and impose no positivity or nonzero-variance condition.

Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by tc · Oct 7, 2026

    Confirmed by the mission captain (proposal self-audit).

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