Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Equation 8.1.2 — Variational trace exponential

Proved
TroppMatrixConcentration.ch8_variational_trace_exp

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

convexitymatrix-analysismatrix-concentration

For every Hermitian complex matrix HHH and positive-definite complex matrix AAA of the same positive dimension,

tr⁡exp⁡(H+log⁡A)=max⁡T≻0{tr⁡(TH)+tr⁡A−D(T;A)}.\operatorname{tr}\exp(H+\log A)=\max_{T\succ0}\{\operatorname{tr}(TH)+\operatorname{tr}A-D(T;A)\}.trexp(H+logA)=T≻0max​{tr(TH)+trA−D(T;A)}.

The maximum is over positive-definite matrices of the same size and is asserted to be attained. This gives the exact variational objective used in the source’s Lieb concavity argument.

Preamble
import Definitions.Def_TroppMatrixConcentration_ch8_entropy

open scoped Matrix.Norms.L2Operator ComplexOrder

Formal statement
namespace TroppMatrixConcentration

theorem ch8_variational_trace_exp {d : ℕ} [NeZero d]
    (H A : Matrix (Fin d) (Fin d) ℂ) (hH : H.IsHermitian) (hA : A.PosDef) :
    IsGreatest {r : ℝ | ∃ T : Matrix (Fin d) (Fin d) ℂ,
      T.PosDef ∧ r = (Matrix.trace (T * H)).re + (Matrix.trace A).re -
        ch8_relativeEntropy T A}
      (traceExp (H + matrixLog A)) := 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; Equation (8.1.2), together with the attainment in Lemma 8.1.6, printed p. 121.
Read-back

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

For every nonzero natural number ddd, every complex d×dd\times dd×d matrix HHH that is Hermitian, and every complex d×dd\times dd×d positive definite matrix AAA, define log⁡B\log BlogB to be the continuous functional calculus of the real logarithm applied to a matrix BBB, and let eBe^BeB denote the matrix exponential. Then the real number Re⁡tr⁡(eH+log⁡A)\operatorname{Re}\operatorname{tr}(e^{H+\log A})Retr(eH+logA) is the greatest element of the set

{Re⁡tr⁡(TH)+Re⁡tr⁡(A)−Re⁡tr⁡ ⁣(T(log⁡T−log⁡A)−(T−A)): T∈Cd×d, T positive definite}.\left\{\operatorname{Re}\operatorname{tr}(TH)+\operatorname{Re}\operatorname{tr}(A)-\operatorname{Re}\operatorname{tr}\!\left(T(\log T-\log A)-(T-A)\right):\ T\in\mathbb C^{d\times d},\ T\text{ positive definite}\right\}.{Retr(TH)+Retr(A)−Retr(T(logT−logA)−(T−A)): T∈Cd×d, T positive definite}.

In particular, every positive definite TTT gives a displayed value at most Re⁡tr⁡(eH+log⁡A)\operatorname{Re}\operatorname{tr}(e^{H+\log A})Retr(eH+logA), and at least one such TTT gives equality; uniqueness is not asserted. Positive definiteness here includes being Hermitian and having strictly positive quadratic form on each nonzero complex vector. The dimension-zero case is excluded, and the matrices over which the greatest value is attained are required to be positive definite, so singular positive semidefinite matrices are excluded.

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