Theorem 6.1.1 — Matrix Bernstein for rectangular matrices
ProvedTroppMatrixConcentration.matrix_bernsteinLet be independent measurable complex random matrices on a probability space, with . Assume and almost surely, where and the norm is the spectral norm. Set . Then
At a zero denominator the tail right-hand side is for and zero for . The finite family may be empty. All moments needed in this statement exist by boundedness. This is the mission's goal, retaining both the expectation and tail conclusions.
import Definitions.Def_TroppMatrixConcentration_probability import Definitions.Def_TroppMatrixConcentration_dilation import Mathlib.Analysis.Convex.Function open MeasureTheory ProbabilityTheory open scoped Matrix.Norms.L2Operator
namespace TroppMatrixConcentration
theorem matrix_bernstein {Ω : Type*} [MeasurableSpace Ω]
(μ : Measure Ω) [IsProbabilityMeasure μ] {m n N : ℕ} [NeZero m] [NeZero n]
(S : Fin N → Ω → Matrix (Fin m) (Fin n) ℂ) (L : ℝ) (hL : 0 ≤ L)
(hMeas : ∀ k, Measurable (S k)) (hIndep : iIndepFun S μ)
(hMean : ∀ k, (∫ ω, S k ω ∂μ) = 0)
(hBound : ∀ k, ∀ᵐ ω ∂μ, spectralNorm (S k ω) ≤ L) :
let Z := fun ω => ∑ k, S k ω
let v := rectSecondMoment μ Z
v = max (spectralNorm (∑ k, ∫ ω, S k ω * (S k ω).conjTranspose ∂μ))
(spectralNorm (∑ k, ∫ ω, (S k ω).conjTranspose * S k ω ∂μ)) ∧
(∫ ω, spectralNorm (Z ω) ∂μ) ≤
Real.sqrt (2 * v * Real.log (m + n)) + L * Real.log (m + n) / 3 ∧
∀ t : ℝ, 0 ≤ t → (μ {ω | t ≤ spectralNorm (Z ω)}).toReal ≤
bernsteinTail (m + n) v L t := by sorry
end TroppMatrixConcentration
Read-back
What the Lean code literally says, in plain math · gpt-6
Let be any measurable space, let be a probability measure on it, let be positive integers and any nonnegative integer, and let , indexed by , be measurable and jointly independent with respect to . Write for conjugate transpose and for the operator norm induced by the Euclidean norms on the complex coordinate spaces, and interpret all matrix-valued integrals as Bochner integrals. Suppose satisfies , each , and, for every , for -almost every . Define and . Then all three of the following hold: ; , where is the natural logarithm; and for every real , , with the probability viewed as a real number and with defined to be if and , to be if and , and otherwise to be . The quantifiers include , , and ; when the sums are empty, and , and the assumptions indexed by are vacuous. In particular is included, its event is all of , and its asserted upper bound is in every case; zero values of are governed by the stated branches rather than division by zero.
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.