Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Finite amplitude dQNTK response and bound

Open
QAlgorithms.Papers.Independent.Intakenext100.P88.amplitude_response

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

prove2quantum

Use the same ordered unitary circuit and normalized feature states, but output all complex amplitudes zx,i=⟨i∣Uϕ(x)⟩z_{x,i}=\langle i|U\phi(x)\ranglezx,i​=⟨i∣Uϕ(x)⟩, i=1,…,hi=1,\ldots,hi=1,…,h. Duplicate each output as zx,0,i=zx,i\mathcal z_{x,0,i}=z_{x,i}zx,0,i​=zx,i​ and zx,1,i=zx,i‾\mathcal z_{x,1,i}=\overline{z_{x,i}}zx,1,i​=zx,i​​. The worldsheet index therefore records whether an amplitude is conjugated. For real parameters, write Jμℓ=∂θℓzμ\mathcal J_{\mu\ell}=\partial_{\theta_\ell}\mathcal z_\muJμℓ​=∂θℓ​​zμ​.

Absorb the reference angles, rescale by δ>0\delta>0δ>0, and compute doubled jets JμℓJ_{\mu\ell}Jμℓ​ and Hμ,kℓH_{\mu,k\ell}Hμ,kℓ​ at zero directly from the actual circuit. For initial angles vvv, set Vμℓ=∑kHμ,kℓvkV_{\mu\ell}=\sum_k H_{\mu,k\ell}v_kVμℓ​=∑k​Hμ,kℓ​vk​, K=JJ†K=JJ^\daggerK=JJ†, and D=JV†+VJ†D=JV^\dagger+VJ^\daggerD=JV†+VJ†, all restricted to doubled training indices. Let eee be the quadratic Taylor amplitude output minus the complex labels and their conjugates. With q=η/2>0q=\eta/2>0q=η/2>0 and B=1−qKB=1-qKB=1−qK, the retained coefficient recurrences Ft+1=BFtF_{t+1}=BF_tFt+1​=BFt​ and It+1=BIt−qDFtI_{t+1}=BI_t-qDF_tIt+1​=BIt​−qDFt​, initialized by (F0,I0)=(e,0)(F_0,I_0)=(e,0)(F0​,I0​)=(e,0), satisfy

Ft=Bte,I0=0,It+1=−q∑s=0tBt−sDBse,∥It+1∥2≤q(t+1)∥B∥2t∥D∥2∥e∥2.F_t=B^t e,\qquad I_0=0,\qquad I_{t+1}=-q\sum_{s=0}^{t}B^{t-s}DB^s e,\qquad \|I_{t+1}\|_2\le q(t+1)\|B\|_2^t\|D\|_2\|e\|_2.Ft​=Bte,I0​=0,It+1​=−qs=0∑t​Bt−sDBse,∥It+1​∥2​≤q(t+1)∥B∥2t​∥D∥2​∥e∥2​.

The retained residual is Ft+ItF_t+I_tFt​+It​. This is a fixed-time coefficient response and requires no impossible full doubled inverse or nonlinear remainder estimate.

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.33 Eqs.(170)-(173): finite doubled-amplitude response and L2 norm bound.
All jets come from the original U; the response retains conjugation symmetry and eta/2.
The coefficient recurrence needs no inverse of the necessarily singular doubled kernel. -/
theorem amplitude_response {D T : Type*} [Fintype T] [DecidableEq T] {h L : ℕ}
    (M : AmplitudeModel D h L) (embed : T ↪ D)
    (delta eta : ℝ) (hdelta : 0 < delta) (heta : 0 < eta)
    (v : Fin L → ℝ) (y : T × Fin h → ℂ) :
    let incl := fun a : T × Bool × Fin h => (embed a.1, a.2.1, a.2.2)
    let K : Matrix (T × Bool × Fin h) (T × Bool × Fin h) ℂ := fun a b =>
      frozenAmplitudeKernel M delta (incl a) (incl b)
    let D : Matrix (T × Bool × Fin h) (T × Bool × Fin h) ℂ := fun a b =>
      deltaAmplitudeKernel M delta v (incl a) (incl b)
    let e : EuclideanSpace ℂ (T × Bool × Fin h) := WithLp.toLp 2 (fun a =>
      quadraticAmplitude M delta v (incl a) -
        (if a.2.1 then star (y (a.1, a.2.2)) else y (a.1, a.2.2)))
    let R := retainedResidual K D (eta / 2) e
    (∀ t, (R t).1 = l2Action ((1 - (eta / 2) • K) ^ t) e) ∧
    (R 0).2 = 0 ∧
    (∀ t : ℕ, (R (t + 1)).2 = -(eta / 2 : ℝ) •
      (∑ s ∈ Finset.range (t + 1),
        l2Action (((1 - (eta / 2) • K) ^ (t - s)) * D * ((1 - (eta / 2) • K) ^ s)) e)) ∧
    ∀ t : ℕ, ‖(R (t + 1)).2‖ ≤ (eta / 2) * ((t : ℝ) + 1) *
      QAlgorithms.specNorm (1 - (eta / 2) • K) ^ t * QAlgorithms.specNorm D * ‖e‖ := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P88

end
Source
Representation Learning via Quantum Neural Tangent Kernels, arXiv:2111.04225v3, PDF p. 33, SM Eqs.(170)-(173)
Read-back

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

Let D,TD,TD,T be types with TTT finite and with decidable equality, 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 XlX_lXl​, and unit vectors ϕx\phi_xϕx​. Define 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, Ax,false,j(θ)=(U(θ)ϕx)jA_{x,\mathrm{false},j}(\theta)=(U(\theta)\phi_x)_jAx,false,j​(θ)=(U(θ)ϕx​)j​, and Ax,true,j(θ)=(U(θ)ϕx)j‾A_{x,\mathrm{true},j}(\theta)=\overline{(U(\theta)\phi_x)_j}Ax,true,j​(θ)=(U(θ)ϕx​)j​​. Fix an injection ι:T↪D\iota:T\hookrightarrow Dι:T↪D, δ,η>0\delta,\eta>0δ,η>0, v∈RLv\in\mathbb R^Lv∈RL, and complex labels y:T×{0,…,h−1}→Cy:T\times\{0,\ldots,h-1\}\to\mathbb Cy:T×{0,…,h−1}→C. Let B=T×{false,true}×{0,…,h−1}B=T\times\{\mathrm{false},\mathrm{true}\}\times\{0,\ldots,h-1\}B=T×{false,true}×{0,…,h−1}, (x,σ,j)^=(ι(x),σ,j)\widehat{(x,\sigma,j)}=(\iota(x),\sigma,j)(x,σ,j)​=(ι(x),σ,j), Jl(a)=∂ulAa(δu)∣u=0J_l(a)=\left.\partial_{u_l}A_a(\delta u)\right|_{u=0}Jl​(a)=∂ul​​Aa​(δu)∣u=0​, and Hkl(a)=∂uk∂ulAa(δu)∣u=0H_{kl}(a)=\left.\partial_{u_k}\partial_{u_l}A_a(\delta u)\right|_{u=0}Hkl​(a)=∂uk​​∂ul​​Aa​(δu)∣u=0​, using real coordinate derivatives. Define Q(a)=Aa(0)+∑lJl(a)vl+12∑k,lHkl(a)vkvlQ(a)=A_a(0)+\sum_lJ_l(a)v_l+\frac12\sum_{k,l}H_{kl}(a)v_kv_lQ(a)=Aa​(0)+∑l​Jl​(a)vl​+21​∑k,l​Hkl​(a)vk​vl​, Kbc=∑lJl(b^)Jl(c^)‾K_{bc}=\sum_lJ_l(\widehat b)\overline{J_l(\widehat c)}Kbc​=∑l​Jl​(b)Jl​(c)​, and

Dbc=∑l(Jl(b^)∑kvkHkl(c^)‾+(∑kvkHkl(b^))Jl(c^)‾).D_{bc}=\sum_l\left( J_l(\widehat b)\overline{\sum_kv_kH_{kl}(\widehat c)} +\left(\sum_kv_kH_{kl}(\widehat b)\right)\overline{J_l(\widehat c)} \right).Dbc​=l∑​(Jl​(b)k∑​vk​Hkl​(c)​+(k∑​vk​Hkl​(b))Jl​(c)​).

Set Yx,false,j=yx,jY_{x,\mathrm{false},j}=y_{x,j}Yx,false,j​=yx,j​, Yx,true,j=yx,j‾Y_{x,\mathrm{true},j}=\overline{y_{x,j}}Yx,true,j​=yx,j​​, eb=Q(b^)−Ybe_b=Q(\widehat b)-Y_beb​=Q(b)−Yb​, and A=IB−η2KA=I_B-\frac{\eta}{2}KA=IB​−2η​K. For complex Euclidean vectors defined by p0=ep_0=ep0​=e, q0=0q_0=0q0​=0, pt+1=Aptp_{t+1}=Ap_tpt+1​=Apt​, and qt+1=Aqt−η2Dptq_{t+1}=Aq_t-\frac{\eta}{2}Dp_tqt+1​=Aqt​−2η​Dpt​, one has pt=Atep_t=A^tept​=Ate for every t∈Nt\in\mathbb Nt∈N, q0=0q_0=0q0​=0, and

qt+1=−η2∑s=0tAt−sDAse,∥qt+1∥2≤η2(t+1)∥A∥opt∥D∥op∥e∥2(t∈N).q_{t+1}=-\frac{\eta}{2}\sum_{s=0}^{t}A^{t-s}DA^se, \qquad \|q_{t+1}\|_2\le\frac{\eta}{2}(t+1)\|A\|_{\mathrm{op}}^t\|D\|_{\mathrm{op}}\|e\|_2 \quad(t\in\mathbb N).qt+1​=−2η​s=0∑t​At−sDAse,∥qt+1​∥2​≤2η​(t+1)∥A∥opt​∥D∥op​∥e∥2​(t∈N).

The operator norms are induced by complex Euclidean norms. Empty TTT and L=0L=0L=0 are allowed, with empty sums equal to zero and empty products equal to the identity; every exponent t−st-st−s in the sum is nonnegative.

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