Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 8.1.6 — Attained variational formula for trace

Proved
TroppMatrixConcentration.ch8_variational_trace

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

convexitymatrix-analysismatrix-concentration

For every positive-definite complex matrix MMM of positive dimension,

tr⁡M=max⁡T≻0tr⁡[Tlog⁡M−Tlog⁡T+T].\operatorname{tr}M=\max_{T\succ0}\operatorname{tr}[T\log M-T\log T+T].trM=T≻0max​tr[TlogM−TlogT+T].

The feasible matrices TTT have the same size as MMM. The assertion includes that every feasible value is bounded above by the displayed trace and that this value is attained. It is the source’s supremum formula with the attainment established in its proof retained explicitly.

Preamble
import Definitions.Def_TroppMatrixConcentration_ch8_entropy

open scoped Matrix.Norms.L2Operator ComplexOrder

Formal statement
namespace TroppMatrixConcentration

theorem ch8_variational_trace {d : ℕ} [NeZero d]
    (M : Matrix (Fin d) (Fin d) ℂ) (hM : M.PosDef) :
    IsGreatest {r : ℝ | ∃ T : Matrix (Fin d) (Fin d) ℂ,
      T.PosDef ∧ r = (Matrix.trace (T * matrixLog M - T * matrixLog T + T)).re}
      (Matrix.trace M).re := 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; Lemma 8.1.6 and its attainment observation, printed p. 121.
Read-back

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

For every natural number d≠0d\ne 0d=0 and every positive-definite complex d×dd\times dd×d matrix MMM, the real number Re⁡tr⁡M\operatorname{Re}\operatorname{tr}MRetrM is the greatest element of the set {Re⁡tr⁡(Tlog⁡M−Tlog⁡T+T):T is a positive-definite complex d×d matrix}\{\operatorname{Re}\operatorname{tr}(T\log M-T\log T+T):T\text{ is a positive-definite complex }d\times d\text{ matrix}\}{Retr(TlogM−TlogT+T):T is a positive-definite complex d×d matrix}. Here positive definiteness includes being Hermitian and having strictly positive quadratic form on every nonzero complex vector, and log⁡A\log AlogA denotes the matrix obtained by applying the real natural logarithm to AAA through continuous functional calculus (equivalently, applying the logarithm to its positive eigenvalues in this setting). Explicitly, some such matrix TTT attains Re⁡tr⁡M\operatorname{Re}\operatorname{tr}MRetrM, and every such matrix TTT satisfies Re⁡tr⁡(Tlog⁡M−Tlog⁡T+T)≤Re⁡tr⁡M\operatorname{Re}\operatorname{tr}(T\log M-T\log T+T)\le\operatorname{Re}\operatorname{tr}MRetr(TlogM−TlogT+T)≤RetrM. The dimension-zero case is excluded; neither uniqueness of an attaining matrix nor a prescribed attaining matrix is asserted.

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