Fourier coefficients of (1 − β|z|/m)^T are nonnegative and small
ProvedQAlgorithms.Arunachalam.fourier_weightPow_boundarunachalamlecture-notesquantum-information
Let , and let be an integer with (the setting of Theorem 7.5.1). Let , , with Fourier coefficients . Then for every of Hamming weight ,
Combined with Claim 7.5.3 this bounds the diagonal of in Theorem 7.5.1.
Formalization Note The conditions and are those of Theorem 7.5.1, which the proof of the lemma uses.
Preamble
import Definitions.Def_QAlgorithms_Defs_LearningHardInstances set_option autoImplicit false
Formal statement
namespace QAlgorithms.Arunachalam
open QAlgorithms.Learning
/-- Arunachalam p.172–175, Lemma 7.5.4, in the setting of Theorem 7.5.1 (`m ≥ 10`,
`T ∈ [1, m/(e³β)]` a natural number, which the proof uses): for `β ∈ (0, 1]`, the Fourier
coefficients of `f(z) = (1 − β|z|/m)^T` on `{0,1}^m` satisfy
`0 ≤ f̂(S) ≤ 4e (1 − β/2)^T (Tβ/m)^q e^{22T²β²/m}` for all `S` with `|S| = q`. -/
theorem fourier_weightPow_bound {m T : ℕ} {β : ℝ} (hm : 10 ≤ m) (hβ : 0 < β) (hβ1 : β ≤ 1)
(hT : 1 ≤ T) (hTm : (T : ℝ) ≤ (m : ℝ) / (Real.exp 3 * β)) (S : Fin m → ZMod 2) :
0 ≤ boolFourier (weightPowFun m β T) S ∧
boolFourier (weightPowFun m β T) S ≤
4 * Real.exp 1 * (1 - β / 2) ^ T * ((T : ℝ) * β / (m : ℝ)) ^ hammingNorm S *
Real.exp (22 * (T : ℝ) ^ 2 * β ^ 2 / (m : ℝ)) := by
sorry
end QAlgorithms.Arunachalam
Source
Srinivasan Arunachalam, Quantum Algorithms and Learning Theory — Part II, 2018 thesis, ILLC DS-2018-08, PDF p. 172, Lemma 7.5.4. https://ir.cwi.nl/pub/27536/Thesis_S_Arunachalam
Read-back
What the Lean code literally says, in plain math · claude-opus-5-5
Let and . Assume:
- ;
- ;
- ;
- .
Let be arbitrary. Define on , where is the Hamming weight. Let be its Boolean Fourier transform.
The theorem asserts
where is the Hamming weight of .
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.