Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Data processing of fidelity under the partial trace

Proved
MixedCloning.Optimal.fidelity_le_partialTrace

by dmitry · Nov 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

data-processingfidelityquantum-channels

Let aaa and bbb be finite sets, let m≥0m\ge0m≥0, and consider mmm registers A1B1,…,AmBmA_1B_1,\dots,A_mB_mA1​B1​,…,Am​Bm​ with At≅CaA_t\cong\mathbb C^aAt​≅Ca and Bt≅CbB_t\cong\mathbb C^bBt​≅Cb. Let F(X,Y)=(tr⁡X YX)2F(X,Y)=\big(\operatorname{tr}\sqrt{\sqrt X\,Y\sqrt X}\big)^2F(X,Y)=(trX​YX​​)2 be the squared fidelity. For all positive semidefinite X,YX,YX,Y on (Ca⊗Cb)⊗m(\mathbb C^a\otimes\mathbb C^b)^{\otimes m}(Ca⊗Cb)⊗m,

F(X,Y) ≤ F(tr⁡B1⋯BmX, tr⁡B1⋯BmY).F(X,Y)\ \le\ F\big(\operatorname{tr}_{B_1\cdots B_m}X,\ \operatorname{tr}_{B_1\cdots B_m}Y\big).F(X,Y) ≤ F(trB1​⋯Bm​​X, trB1​⋯Bm​​Y).

This is the data-processing inequality for fidelity [Wat18, Theorem 3.27], specialized to the channel that traces out the auxiliary registers. The source uses it in the proofs of Proposition 10 (p. 16) and Proposition 25 (p. 27) to pass from purified outputs to the original registers.

Formalization Note The partial trace is the platform function partialTrace m, which keeps the register order A1⋯AmA_1\cdots A_mA1​⋯Am​. The inequality is stated for squared fidelity; it is equivalent to the root-fidelity form because root fidelity is nonnegative.

Differences from the source:

  • specialization: an arbitrary channel Φ\PhiΦ → the partial trace over B1,…,BmB_1,\dots,B_mB1​,…,Bm​; reason: this is the instance used in the source
Preamble
import Definitions.Def_MixedCloning_Optimal_Model

noncomputable section
open Matrix MeasureTheory
open scoped Matrix ComplexConjugate ComplexOrder Matrix.Norms.Elementwise
Formal statement
namespace MixedCloning.Optimal
attribute [local instance] Classical.propDecidable Classical.decEq

/-- Data processing of fidelity [Wat18, Theorem 3.27], used on PDF p.16 (and p.27), for the
partial trace over the auxiliary registers `B_1,…,B_m`. -/
theorem fidelity_le_partialTrace {a b : Type} [Fintype a] [DecidableEq a] [Fintype b] [DecidableEq b] (m : ℕ)
    (X Y : Op (Word (a × b) m)) (hX : X.PosSemidef) (hY : Y.PosSemidef) :
    fidelity X Y ≤ fidelity (partialTrace m X) (partialTrace m Y) := by sorry

end MixedCloning.Optimal
Source
John Watrous, The Theory of Quantum Information, Cambridge University Press, 2018, Theorem 3.27 (monotonicity of fidelity under channels), https://doi.org/10.1017/9781316848142; used as 'data processing of fidelity [Wat18, Theorem 3.27]' in Dmitry Grinko, Marco Fanizza, Thilo Scharnhorst, Jack Spilecki, Optimal cloning of mixed states, STOC 2027 submission #243 (October 2026), https://github.com/sigact-stoc/stoc2027/releases/download/papers-1/243.pdf, p. 16 (proof of Proposition 10) and p. 27 (proof of Proposition 25).

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