Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Eq. (33), pp. 11–13 — balanced approximate group commutators in SU(d)

Proved
SKAlgorithm.Qudit.eq_33

by mikedeng1 · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

group-commutatorp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1solovay-kitaevspecial-unitary-group

Fix ddd. There are constants cgc′>0c_{gc'} > 0cgc′​>0 and cgc′′>0c_{gc''} > 0cgc′′​>0 such that for every ϵ>0\epsilon > 0ϵ>0 and every U∈SU(d)U \in SU(d)U∈SU(d) with d(I,U)<ϵd(I, U) < \epsilond(I,U)<ϵ there exist V,W∈SU(d)V, W \in SU(d)V,W∈SU(d) with

d(VWV†W†,U)<cgc′ ϵ3/2,d(I,V), d(I,W)<cgc′′ϵ.d(VWV^\dagger W^\dagger, U) < c_{gc'}\, \epsilon^{3/2}, \qquad d(I, V),\ d(I, W) < c_{gc''} \sqrt{\epsilon} .d(VWV†W†,U)<cgc′​ϵ3/2,d(I,V), d(I,W)<cgc′′​ϵ​.

This is the guarantee of GC-Approx-Decompose in the qudit algorithm: a gate near the identity is approximated, to order ϵ3/2\epsilon^{3/2}ϵ3/2, by a group commutator of two gates at distance O(ϵ)O(\sqrt\epsilon)O(ϵ​) from the identity. The paper's constants are cgc′′≈d1/4((d−1)/2)1/2c_{gc''} \approx d^{1/4}((d-1)/2)^{1/2}cgc′′​≈d1/4((d−1)/2)1/2 and cgc′≈4d3/4((d−1)/2)3/2c_{gc'} \approx 4 d^{3/4}((d-1)/2)^{3/2}cgc′​≈4d3/4((d−1)/2)3/2.

Formalization Note. The constants are existential, chosen before ϵ\epsilonϵ and UUU, and may depend on ddd; the approximate values are not pinned. VVV and WWW are required to lie in SU(d)SU(d)SU(d), since the algorithm recurses on them. The paper's derivation on pp. 12–13 contains a sign slip: with [F,G]=iH[F, G] = iH[F,G]=iH and U=exp⁡(iH)U = \exp(iH)U=exp(iH), Lemma 3 gives VWV†W†≈exp⁡(i⋅i[F,G])=exp⁡(−iH)=U†VWV^\dagger W^\dagger \approx \exp(i \cdot i[F,G]) = \exp(-iH) = U^\daggerVWV†W†≈exp(i⋅i[F,G])=exp(−iH)=U†, not UUU. The statement of Eq. (33) is nevertheless true (apply Lemma 2 to −H-H−H, or exchange FFF and GGG), and it is stated as the paper claims it.

Preamble
import Mathlib
import Definitions.Def_SKAlgorithm_Qudit_Model

open scoped Matrix
open QAlgorithms
Formal statement
namespace SKAlgorithm.Qudit

theorem eq_33 (d : ℕ) :
    ∃ c' c'' : ℝ, 0 < c' ∧ 0 < c'' ∧ ∀ ε : ℝ, 0 < ε → ∀ U : SU d, opDist 1 U < ε →
      ∃ V W : SU d, opDist (V * W * V⁻¹ * W⁻¹) U < c' * ε ^ ((3 : ℝ) / 2) ∧
        opDist 1 V < c'' * Real.sqrt ε ∧ opDist 1 W < c'' * Real.sqrt ε := by sorry

end SKAlgorithm.Qudit
Source
Dawson and Nielsen, The Solovay-Kitaev algorithm, arXiv:quant-ph/0505030v2, p. 11, §5.2 (goal of the subsection); p. 13, Eq. (33); p. 10, §5.1, properties (a)–(b)
Human review
  • Endorsed by Shuze Chen · Oct 22, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 22, 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