Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Low-photon nats expansions and diverging photon efficiency

Open
QAlgorithms.Papers.Independent.Intakenext100.P76.lowPhotonExpansions

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

prove2quantum

Let G(x)=(ln⁡2)g(x)G(x)=(\ln2)g(x)G(x)=(ln2)g(x) and B(x)=(ln⁡2)h2((1−e−2x)/2)B(x)=(\ln2)h_2((1-e^{-2x})/2)B(x)=(ln2)h2​((1−e−2x)/2) denote the two Holevo expressions in nats per mode. As x↓0x\downarrow0x↓0 through strictly positive photon numbers,

G(x)=−xln⁡x+x+x22+O(x3),B(x)=−xln⁡x+x+x2ln⁡x+O(x2).G(x)=-x\ln x+x+\frac{x^2}{2}+O(x^3),\qquad B(x)=-x\ln x+x+x^2\ln x+O(x^2).G(x)=−xlnx+x+2x2​+O(x3),B(x)=−xlnx+x+x2lnx+O(x2).

Explicitly, there are positive constants C,D,δC,D,\deltaC,D,δ, with δ≤1\delta\le1δ≤1, bounding the absolute remainders by Cx3Cx^3Cx3 and Dx2Dx^2Dx2 for every 0<x<δ0<x<\delta0<x<δ. Furthermore,

G(x)x+ln⁡x⟶1,B(x)x+ln⁡x⟶1,G(x)x, B(x)x⟶+∞.\frac{G(x)}x+\ln x\longrightarrow1,\qquad \frac{B(x)}x+\ln x\longrightarrow1,\qquad \frac{G(x)}x,\ \frac{B(x)}x\longrightarrow+\infty.xG(x)​+lnx⟶1,xB(x)​+lnx⟶1,xG(x)​, xB(x)​⟶+∞.

These limits express the common first two terms and the unbounded photon efficiencies.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Intakenext100_P76
Formal statement
noncomputable section

namespace QAlgorithms.Papers.Independent.Intakenext100.P76

open MeasureTheory Filter

open QAlgorithms.Papers.Independent.Information100.P63 (logTwo normalCDF normalQuantile)

open QAlgorithms.Papers.Independent.Information100.P54 (thermalEntropy)

open QAlgorithms.Papers.Independent.Dependency50.P39 (coherentCoefficient)

open QAlgorithms.Papers.Independent.Next46.P32 (PositiveKernel)

open scoped BigOperators ENNReal Topology

attribute [local instance] Classical.propDecidable

attribute [local instance] Classical.decEq

/-- PDF p.13, (6.15)–(6.16) and the adjacent photon-efficiency assertion.
Nats are explicit. Remainders are O(x^3) and O(x^2), respectively, as x goes
to zero from the right. In particular no evaluation at x=0 or negative x is used. -/
theorem lowPhotonExpansions :
    (∃ C D delta : ℝ, 0 < C ∧ 0 < D ∧ 0 < delta ∧ delta ≤ 1 ∧
      ∀ x : ℝ, 0 < x → x < delta →
        |thermalEntropyNats x - (-x * Real.log x + x + x ^ 2 / 2)| ≤ C * x ^ 3 ∧
        |bpskHolevoNats x - (-x * Real.log x + x + x ^ 2 * Real.log x)| ≤ D * x ^ 2) ∧
    Tendsto (fun x : ℝ => thermalEntropyNats x / x + Real.log x) (𝓝[>] 0) (𝓝 1) ∧
    Tendsto (fun x : ℝ => bpskHolevoNats x / x + Real.log x) (𝓝[>] 0) (𝓝 1) ∧
    Tendsto (fun x : ℝ => thermalEntropyNats x / x) (𝓝[>] 0) atTop ∧
    Tendsto (fun x : ℝ => bpskHolevoNats x / x) (𝓝[>] 0) atTop := by sorry

end QAlgorithms.Papers.Independent.Intakenext100.P76

end
Source
Second-order coding rates for pure-loss bosonic channels, arXiv:1408.5328v3, PDF p. 13, §6.2, (6.15)–(6.16) and adjacent efficiency assertion
Read-back

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

Define, for real xxx,

T(x)=(x+1)ln⁡(x+1)−xln⁡x,B(x)=ln⁡2 H2 ⁣(1−e−2x2),H2(t)=−tln⁡t−(1−t)ln⁡(1−t)ln⁡2,T(x)=(x+1)\ln(x+1)-x\ln x, \qquad B(x)=\ln2\,H_2\!\left(\frac{1-e^{-2x}}2\right), \qquad H_2(t)=\frac{-t\ln t-(1-t)\ln(1-t)}{\ln2},T(x)=(x+1)ln(x+1)−xlnx,B(x)=ln2H2​(21−e−2x​),H2​(t)=ln2−tlnt−(1−t)ln(1−t)​,

using the total real logarithm with ln⁡0=0\ln0=0ln0=0. There exist real constants C,D,δC,D,\deltaC,D,δ with C>0C>0C>0, D>0D>0D>0, and 0<δ≤10<\delta\le10<δ≤1, such that every real xxx with 0<x<δ0<x<\delta0<x<δ satisfies both

∣T(x)−(−xln⁡x+x+x22)∣≤Cx3,∣B(x)−(−xln⁡x+x+x2ln⁡x)∣≤Dx2.\left|T(x)-\left(-x\ln x+x+\frac{x^2}{2}\right)\right|\le Cx^3, \qquad \left|B(x)-\left(-x\ln x+x+x^2\ln x\right)\right|\le Dx^2.​T(x)−(−xlnx+x+2x2​)​≤Cx3,​B(x)−(−xlnx+x+x2lnx)​≤Dx2.

In addition, as xxx tends to 000 through strictly positive values,

T(x)x+ln⁡x⟶1,B(x)x+ln⁡x⟶1,T(x)x⟶+∞,B(x)x⟶+∞.\frac{T(x)}x+\ln x\longrightarrow1,\qquad \frac{B(x)}x+\ln x\longrightarrow1,\qquad \frac{T(x)}x\longrightarrow+\infty,\qquad \frac{B(x)}x\longrightarrow+\infty.xT(x)​+lnx⟶1,xB(x)​+lnx⟶1,xT(x)​⟶+∞,xB(x)​⟶+∞.

The error bounds and limits concern x>0x>0x>0, although both functions are also defined at x=0x=0x=0.

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