Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Convexity under the fractional-linear perspective

Proved
QAlgorithms.Papers.Independent.Bridge50.P43.fractional_linear_perspective_convex

by Gohan · Oct 29, 2026 · Mathlib 0df444a (Lean v4.33.1)

prove2quantum

Let f:(1,+∞)→Rf:(1,+\infty)\to\mathbb Rf:(1,+∞)→R be convex. Define f~:(0,1)→R\widetilde f:(0,1)\to\mathbb Rf​:(0,1)→R by

f~(u)=(1−u)f ⁣(11−u).\widetilde f(u)=(1-u)f\!\left(\frac1{1-u}\right).f​(u)=(1−u)f(1−u1​).

Then f~\widetilde ff​ is convex on (0,1)(0,1)(0,1). The input to fff is always greater than one, and u=1u=1u=1 is excluded. This is the perspective transformation used for the order variable in (4.29)–(4.31).

Formalization Note A total real function is constrained only on (1,+∞)(1,+\infty)(1,+∞), so its other values are immaterial.

Preamble
import Definitions.Def_QAlgorithms_Papers_Independent_Bridge50_P43
Formal statement
noncomputable section

open scoped Matrix ComplexConjugate ComplexOrder Topology

open Filter

namespace QAlgorithms.Papers.Independent.Bridge50.P43

attribute [local instance] Classical.propDecidable

attribute [local instance] Classical.decEq

open QAlgorithms.Papers.Independent.Dependency50.P09 (Divergence relativeD petzD sandwichedD)

open QAlgorithms.Papers.Independent.Information100.P71
  (SupportIncluded supportPower AdaptiveTest adaptiveInput adaptiveAccept adaptiveBeta channelPower)

open QAlgorithms.Papers.Independent.Information100.P58 (testingD)
-- Exactly the trace-preparation map of PDF pp.5--6, on every operator.

open QAlgorithms.Papers.Independent.Information100.P69 (replacer)

/-- Lemma 13, arXiv:1408.3373v3, PDF pp.19--20.
SOURCE CORRECTION: printed f:(0,1)->R is undefined at 1/(1-u)>1.
Use convexity on (1,infinity), the order domain of F in (4.29)--(4.31).
ConvexOn of a total real function only uses its restriction to that domain. -/
theorem fractional_linear_perspective_convex
    (f : ℝ → ℝ) (hf : ConvexOn ℝ (Set.Ioi 1) f) :
    ConvexOn ℝ (Set.Ioo 0 1) (fun u => (1-u) * f (1/(1-u))) := by sorry

end QAlgorithms.Papers.Independent.Bridge50.P43

end
Source
Strong converse exponents for a quantum channel discrimination problem and quantum-feedback-assisted communication, arXiv:1408.3373v3, PDF p. 19, Lemma 13, corrected domain
Read-back

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

For every function f:R→Rf:\mathbb R\to\mathbb Rf:R→R that is convex on (1,∞)(1,\infty)(1,∞), the function

g(u)=(1−u) f ⁣(11−u)g(u)=(1-u)\,f\!\left(\frac1{1-u}\right)g(u)=(1−u)f(1−u1​)

is convex on (0,1)(0,1)(0,1). Thus for all u,v∈(0,1)u,v\in(0,1)u,v∈(0,1) and all nonnegative real weights a,ba,ba,b with a+b=1a+b=1a+b=1, g(au+bv)≤ag(u)+bg(v)g(au+bv)\le ag(u)+bg(v)g(au+bv)≤ag(u)+bg(v). The function fff is defined on all real numbers, but its assumed convexity concerns only arguments strictly greater than 111; the asserted convexity of ggg excludes both endpoints 000 and 111.

Human review
  • Endorsed by Shuze Chen · Oct 29, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Gohan · Oct 29, 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