Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 26: unitary differences and the smallest spectral disk

Open
QAlgorithms.Papers.Independent.Intakenext100.P36.unitary_difference_disk

by Gohan · Oct 31, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let n≥1n\ge1n≥1, let U∈MnU\in M_nU∈Mn​ be unitary, and define the complex-linear map Φ(X)=UXU†−X\Phi(X)=UXU^\dagger-XΦ(X)=UXU†−X on all matrices. There exist c∈Cc\in\mathbb Cc∈C and r≥0r\ge0r≥0 such that the closed disk {z:∣z−c∣≤r}\{z:|z-c|\le r\}{z:∣z−c∣≤r} contains the spectrum of UUU and rrr is no greater than the radius of any other closed disk containing that spectrum. For this smallest radius,

∥Φ∥cb=∥Φ†∥cb=∥Φ∥⋄=∥Φ†∥⋄=2r.\|\Phi\|_{\rm cb}=\|\Phi^\dagger\|_{\rm cb}=\|\Phi\|_\diamond=\|\Phi^\dagger\|_\diamond=2r.∥Φ∥cb​=∥Φ†∥cb​=∥Φ∥⋄​=∥Φ†∥⋄​=2r.

The dual is the bilinear trace dual; the diamond norm uses unnormalized trace norm without a factor 1/21/21/2. Scalar unitaries are included, with r=0r=0r=0.

This geometric formula computes the common stabilized norm from a smallest enclosing disk, whose diameter need not equal the maximum distance between two eigenvalues. The paper gives a new elementary proof of a previously known formula.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P36
Formal statement
noncomputable section

open scoped Matrix ComplexConjugate ENNReal ComplexOrder

namespace QAlgorithms.Papers.Independent.Intakenext100.P36

local notation "Mat" => QAlgorithms.Papers.Independent.Information100.P50.M

local notation "Map" => QAlgorithms.Papers.Independent.Information100.P50.Lin

open QAlgorithms.Papers.Independent.Dependency50.P49 (traceDual)

/-- Theorem 26, PDF pp.13–14: four norms equal twice the least enclosing radius.
This paper supplies an elementary proof of an earlier geometric formula. -/
theorem unitary_difference_disk {n : ℕ} (hn : 0 < n)
    (U : Matrix.unitaryGroup (Fin n) ℂ) :
    ∃ center : ℂ, ∃ radius : ℝ,
      IsSmallestSpectralDisk (U : Mat n) center radius ∧
      extendedCBNorm (unitaryDifference (U : Mat n)) = ENNReal.ofReal (2 * radius) ∧
      extendedCBNorm (traceDual (unitaryDifference (U : Mat n))) = ENNReal.ofReal (2 * radius) ∧
      rectangularDiamondNorm (unitaryDifference (U : Mat n)) = ENNReal.ofReal (2 * radius) ∧
      rectangularDiamondNorm (traceDual (unitaryDifference (U : Mat n))) =
        ENNReal.ofReal (2 * radius) := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P36

end
Source
Computing Stabilized Norms for Quantum Operations via the Theory of Completely Bounded Maps, arXiv:0711.3636v1, PDF p. 13, Theorem 26
Read-back

What the Lean code literally says, in plain math · gpt-6.1-sol

Let nnn be a positive natural number and UUU a complex unitary n×nn\times nn×n matrix. There exist z0∈Cz_0\in\mathbb Cz0​∈C and r∈Rr\in\mathbb Rr∈R such that r≥0r\ge0r≥0, ∣z−z0∣≤r|z-z_0|\le r∣z−z0​∣≤r for every zzz in the complex spectrum of UUU, and r≤sr\le sr≤s for every nonnegative sss and center c∈Cc\in\mathbb Cc∈C whose closed disk contains that spectrum. Define ΔU(X)=UXU†−X\Delta_U(X)=UXU^\dagger-XΔU​(X)=UXU†−X, and let ΔU∨\Delta_U^\veeΔU∨​ be its trace dual, characterized by tr⁡(ΔU(X)Y)=tr⁡(XΔU∨(Y))\operatorname{tr}(\Delta_U(X)Y)=\operatorname{tr}(X\Delta_U^\vee(Y))tr(ΔU​(X)Y)=tr(XΔU∨​(Y)); explicitly, ΔU∨(Y)=U†YU−Y\Delta_U^\vee(Y)=U^\dagger YU-YΔU∨​(Y)=U†YU−Y. For a complex linear map ψ:Mn(C)→Mn(C)\psi:M_n(\mathbb C)\to M_n(\mathbb C)ψ:Mn​(C)→Mn​(C), set

C(ψ)=[sup⁡m∈N, X∈Mmn(C)∥X∥op≤1∥(idm⊗ψ)(X)∥op]+,D(ψ)=sup⁡X∈Mn2(C)∥X∥1≤1[∥(idn⊗ψ)(X)∥1]+,C(\psi)= \left[\sup_{\substack{m\in\mathbb N,\ X\in M_{mn}(\mathbb C)\\\|X\|_{\mathrm{op}}\le1}} \|(\mathrm{id}_m\otimes\psi)(X)\|_{\mathrm{op}}\right]_+, \qquad D(\psi)= \sup_{\substack{X\in M_{n^2}(\mathbb C)\\\|X\|_1\le1}} [\|(\mathrm{id}_n\otimes\psi)(X)\|_1]_+,C(ψ)=​m∈N, X∈Mmn​(C)∥X∥op​≤1​sup​∥(idm​⊗ψ)(X)∥op​​+​,D(ψ)=X∈Mn2​(C)∥X∥1​≤1​sup​[∥(idn​⊗ψ)(X)∥1​]+​,

where the supremum inside CCC is real, [t]+[t]_+[t]+​ sends ttt to max⁡(t,0)\max(t,0)max(t,0) in [0,∞][0,\infty][0,∞], amplification acts on matrix blocks, ∥⋅∥op\|\cdot\|_{\mathrm{op}}∥⋅∥op​ is the Euclidean operator norm, and ∥X∥1=Re⁡tr⁡X†X\|X\|_1=\operatorname{Re}\operatorname{tr}\sqrt{X^\dagger X}∥X∥1​=RetrX†X​. For the same rrr, all four equalities hold:

C(ΔU)=C(ΔU∨)=D(ΔU)=D(ΔU∨)=[2r]+.C(\Delta_U)=C(\Delta_U^\vee) =D(\Delta_U)=D(\Delta_U^\vee)=[2r]_+.C(ΔU​)=C(ΔU∨​)=D(ΔU​)=D(ΔU∨​)=[2r]+​.

The radius may be zero, and no uniqueness of the chosen center or radius is asserted.

Human review
  • Endorsed by Shuze Chen · Oct 31, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 31, 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