Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Finite frozen scalar optimization solution

Proved
QAlgorithms.Papers.Independent.Intakenext100.P88.frozen_optimization_finite

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

prove2quantum

The circuit acts on a nonzero finite-dimensional complex Hilbert space H=ChH=\mathbb C^hH=Ch and has the ordered form U(θ)=∏ℓ=1LWℓexp⁡(iθℓXℓ)U(\theta)=\prod_{\ell=1}^L W_\ell\exp(i\theta_\ell X_\ell)U(θ)=∏ℓ=1L​Wℓ​exp(iθℓ​Xℓ​), with fixed unitary WℓW_\ellWℓ​, Hermitian XℓX_\ellXℓ​, and real angles. Products are written in increasing index order and the rightmost factor acts first. Feature vectors ϕ(x)\phi(x)ϕ(x) have norm one. For Hermitian observables OiO_iOi​, put zx,i(θ)=⟨U(θ)ϕ(x),OiU(θ)ϕ(x)⟩∈Rz_{x,i}(\theta)=\langle U(\theta)\phi(x),O_iU(\theta)\phi(x)\rangle\in\mathbb Rzx,i​(θ)=⟨U(θ)ϕ(x),Oi​U(θ)ϕ(x)⟩∈R.

Absorb the reference angles into the fixed gates and set θ=δφ\theta=\delta\varphiθ=δφ, where δ>0\delta>0δ>0. Write Jaℓ=∂φℓza(0)J_{a\ell}=\partial_{\varphi_\ell}z_a(0)Jaℓ​=∂φℓ​​za​(0) and Ha,kℓ=∂φk∂φℓza(0)H_{a,k\ell}=\partial_{\varphi_k}\partial_{\varphi_\ell}z_a(0)Ha,kℓ​=∂φk​​∂φℓ​​za​(0) for derivatives of the actual circuit. The frozen kernel is Kab=∑ℓJaℓJbℓK_{ab}=\sum_\ell J_{a\ell}J_{b\ell}Kab​=∑ℓ​Jaℓ​Jbℓ​.

For scalar optimization, let O0O_0O0​ be the known minimum eigenvalue of OOO and let e=z(0)+∑ℓJℓvℓ−O0e=z(0)+\sum_\ell J_\ell v_\ell-O_0e=z(0)+∑ℓ​Jℓ​vℓ​−O0​ be the initial frozen residual at rescaled angles vvv. Under the frozen recurrence ϵt+1=ϵt−ηKϵt\epsilon_{t+1}=\epsilon_t-\eta K\epsilon_tϵt+1​=ϵt​−ηKϵt​, with η>0\eta>0η>0, one has

ϵt=(1−ηK)te,t∈N.\epsilon_t=(1-\eta K)^t e,\qquad t\in\mathbb N.ϵt​=(1−ηK)te,t∈N.

The identity permits K=0K=0K=0, when the residual stays constant. It is the independently displayed finite solution, without the invalid unconditional convergence assertion of the full printed Theorem 1.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P88
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P88

open scoped BigOperators ComplexConjugate Topology NNReal ComplexOrder

open MeasureTheory ProbabilityTheory Filter

/-- PDF p.3 Eq.(10), SM p.21 Eq.(69): finite frozen scalar solution.
This independent displayed identity does not include Theorem 1's invalid unconditional
zero-limit assertion. K=0 and L=0 are allowed and give constant residuals. -/
theorem frozen_optimization_finite {h L : ℕ} (M : OptimizationModel h L)
    (delta eta : ℝ) (hdelta : 0 < delta) (heta : 0 < eta)
    (v : Fin L → ℝ) (t : ℕ) :
    let K := frozenKernel M.model delta ((), ()) ((), ())
    let e := linearOutput M.model delta v ((), ()) - M.minimum
    frozenResidual (fun _ _ : Unit => K) eta (WithLp.toLp 2 (fun _ : Unit => e)) t () =
      (1 - eta * K) ^ t * e := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P88

end
Source
Representation Learning via Quantum Neural Tangent Kernels, arXiv:2111.04225v3, PDF p. 21, SM Eq.(69); main Eq.(10)
Read-back

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

Let h,Lh,Lh,L be natural numbers and let MMM consist of h>0h>0h>0, unitary matrices Wl∈Ch×hW_l\in\mathbb C^{h\times h}Wl​∈Ch×h, Hermitian matrices Xl∈Ch×hX_l\in\mathbb C^{h\times h}Xl​∈Ch×h, a unit vector ϕ∈Ch\phi\in\mathbb C^hϕ∈Ch, a Hermitian observable OOO, and a real number m∗m_*m∗​. Assume that some unit vector www satisfies Ow=m∗wOw=m_*wOw=m∗​w, and that m∗≤Re⁡⟨w,Ow⟩m_*\le\operatorname{Re}\langle w,Ow\ranglem∗​≤Re⟨w,Ow⟩ for every unit vector w∈Chw\in\mathbb C^hw∈Ch. Define f(θ)=Re⁡⟨U(θ)ϕ,OU(θ)ϕ⟩f(\theta)=\operatorname{Re}\langle U(\theta)\phi,OU(\theta)\phi\ranglef(θ)=Re⟨U(θ)ϕ,OU(θ)ϕ⟩, where U(θ)=∏l=0L−1Wlexp⁡(iθlXl)U(\theta)=\prod_{l=0}^{L-1}W_l\exp(i\theta_lX_l)U(θ)=∏l=0L−1​Wl​exp(iθl​Xl​) in increasing index order. For arbitrary δ,η>0\delta,\eta>0δ,η>0, v∈RLv\in\mathbb R^Lv∈RL, and t∈Nt\in\mathbb Nt∈N, let Jl=ddsf(δs el)∣s=0J_l=\left.\frac{d}{ds}f(\delta s\,e_l)\right|_{s=0}Jl​=dsd​f(δsel​)​s=0​, K=∑lJl2K=\sum_lJ_l^2K=∑l​Jl2​, and e=f(0)+∑lJlvl−m∗e=f(0)+\sum_lJ_lv_l-m_*e=f(0)+∑l​Jl​vl​−m∗​, where ele_lel​ denotes the lll-th coordinate vector. On the one-point index set, start with the vector having sole coordinate eee, and repeatedly multiply by the matrix whose sole entry is 1−ηK1-\eta K1−ηK. The sole coordinate after ttt steps is exactly (1−ηK)te(1-\eta K)^t e(1−ηK)te. The statement includes t=0t=0t=0, with the zeroth power equal to 111, and L=0L=0L=0, when K=0K=0K=0 and the parameter sums are empty.

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