Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 7.7 — convolution on L¹(ℝⁿ), Young's inequality, (f ∗ g)^ = (2π)^{n/2} f̂ ĝ

Proved
TeschlQM.Free.fourier_convolution

by mikedeng1 · Sep 29, 2026 · Mathlib 0df444a (Lean v4.33.1)

fourier-transformharmonic-analysisp2o-batch-books5p2o-gran-per-chapterp2o-plan-bookp2o-v1

For f,g∈L1(Rn)f, g \in L^1(\mathbb R^n)f,g∈L1(Rn) the convolution

(f∗g)(x)=∫Rnf(y)g(x−y) dny=∫Rnf(x−y)g(y) dny(f * g)(x) = \int_{\mathbb R^n} f(y) g(x - y)\, d^ny = \int_{\mathbb R^n} f(x - y) g(y)\, d^ny(f∗g)(x)=∫Rn​f(y)g(x−y)dny=∫Rn​f(x−y)g(y)dny

is again in L1(Rn)L^1(\mathbb R^n)L1(Rn), satisfies Young's inequality ∥f∗g∥1≤∥f∥1∥g∥1\|f * g\|_1 \le \|f\|_1 \|g\|_1∥f∗g∥1​≤∥f∥1​∥g∥1​, and its Fourier transform (normalization f^(p)=(2π)−n/2∫e−ipxf(x) dnx\hat f(p) = (2\pi)^{-n/2}\int e^{-ipx} f(x)\,d^nxf^​(p)=(2π)−n/2∫e−ipxf(x)dnx) is

(f∗g)∧(p)=(2π)n/2f^(p)g^(p).(f * g)^\wedge(p) = (2\pi)^{n/2} \hat f(p) \hat g(p).(f∗g)∧(p)=(2π)n/2f^​(p)g^​(p).

The convolution formula is how the free time evolution is written as an integral operator.

Formalization Note. The convolution is the explicit integral x↦∫f(y)g(x−y) dyx \mapsto \int f(y) g(x-y)\,dyx↦∫f(y)g(x−y)dy; at the null set of xxx where the integrand is not integrable the Bochner integral is 000, which changes neither the L1L^1L1 norm nor the Fourier transform. The two expressions in (7.17) are asserted equal for every xxx.

Preamble
import Mathlib
import Definitions.Def_TeschlQM_Free_fourier
Formal statement
namespace TeschlQM.Free

open MeasureTheory

/-- Teschl, Lemma 7.7, p. 165: for `f, g ∈ L¹(ℝⁿ)` the convolution
`(f ∗ g)(x) = ∫ f(y) g(x - y) dⁿy = ∫ f(x - y) g(y) dⁿy` (7.17) is in `L¹(ℝⁿ)`, satisfies Young's
inequality `‖f ∗ g‖₁ ≤ ‖f‖₁ ‖g‖₁` (7.18), and `(f ∗ g)^(p) = (2π)^{n/2} f̂(p) ĝ(p)` (7.19) with
Teschl's normalization (7.3). -/
theorem fourier_convolution (n : ℕ) (f g : EuclideanSpace ℝ (Fin n) → ℂ)
    (hf : Integrable f volume) (hg : Integrable g volume) :
    (∀ x, ∫ y, f y * g (x - y) = ∫ y, f (x - y) * g y) ∧
      Integrable (fun x => ∫ y, f y * g (x - y)) volume ∧
      ∫ x, ‖∫ y, f y * g (x - y)‖ ≤ (∫ x, ‖f x‖) * ∫ x, ‖g x‖ ∧
      fourier n (fun x => ∫ y, f y * g (x - y)) =
        fun p => (((2 * Real.pi) ^ ((n : ℝ) / 2) : ℝ) : ℂ) * fourier n f p * fourier n g p := by sorry

end TeschlQM.Free
Source
Teschl, Mathematical Methods in Quantum Mechanics, AMS GSM 99, 2009, p. 165, Lemma 7.7
Read-back

What the Lean code literally says, in plain math · claude-opus-5-5

Let n∈Nn\in\mathbb Nn∈N and let f,g:Rn→Cf,g:\mathbb R^n\to\mathbb Cf,g:Rn→C both be Lebesgue integrable on Euclidean Rn\mathbb R^nRn. For every xxx, define

h(x)=∫Rnf(y) g(x−y) dy.h(x)=\int_{\mathbb R^n} f(y)\,g(x-y)\,dy .h(x)=∫Rn​f(y)g(x−y)dy.

This is a Bochner integral, taken to be 000 at any xxx where y↦f(y)g(x−y)y\mapsto f(y)g(x-y)y↦f(y)g(x−y) is not integrable. For any function uuu, write

u^(p)=1(2π)n/2∫e−i⟨p,x⟩u(x) dx,\widehat u(p)=\frac{1}{(2\pi)^{n/2}}\int e^{-i\langle p,x\rangle}u(x)\,dx ,u(p)=(2π)n/21​∫e−i⟨p,x⟩u(x)dx,

with the same zero convention. The statement asserts all four of the following.

  1. For every x∈Rnx\in\mathbb R^nx∈Rn,
∫f(y) g(x−y) dy=∫f(x−y) g(y) dy.\int f(y)\,g(x-y)\,dy=\int f(x-y)\,g(y)\,dy .∫f(y)g(x−y)dy=∫f(x−y)g(y)dy.

Both sides use the zero convention at points where their integrands are not integrable. 2. hhh is Lebesgue integrable. 3. The L1L^1L1 bound holds:

∫Rn∣h(x)∣ dx  ≤  (∫∣f(x)∣ dx)(∫∣g(x)∣ dx).\int_{\mathbb R^n}|h(x)|\,dx\;\le\;\Bigl(\int|f(x)|\,dx\Bigr)\Bigl(\int|g(x)|\,dx\Bigr).∫Rn​∣h(x)∣dx≤(∫∣f(x)∣dx)(∫∣g(x)∣dx).
  1. For every ppp,
h^(p)=(2π)n/2 f^(p) g^(p).\widehat h(p)=(2\pi)^{n/2}\,\widehat f(p)\,\widehat g(p).h(p)=(2π)n/2f​(p)g​(p).

Degenerate cases. When n=0n=0n=0, R0\mathbb R^0R0 is one point of measure 111, so h(0)=f(0)g(0)h(0)=f(0)g(0)h(0)=f(0)g(0). Clause 1 is then trivial, clause 3 reads ∣f(0)g(0)∣≤∣f(0)∣ ∣g(0)∣|f(0)g(0)|\le|f(0)|\,|g(0)|∣f(0)g(0)∣≤∣f(0)∣∣g(0)∣, and clause 4 reads f(0)g(0)=1⋅f(0)g(0)f(0)g(0)=1\cdot f(0)g(0)f(0)g(0)=1⋅f(0)g(0). Non-integrable fff or ggg are excluded by hypothesis.

Human review
  • Endorsed by Shuze Chen · Oct 2, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 2, 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