Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§3.1, proof of Theorem 3.1, (8), p. 484 — for x* ∈ dom h, ∂_εh(x*) ⊂ ∂_εg(x*) ∀ε > 0 is condition (8)

Proved
TaoAnDCA.GlobalOpt.epsSubdiff_subset_iff_eq8

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

conjugatedc-programmingepsilon-subdifferentialp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let g,h∈Γ0(Rn)g,h\in\Gamma_0(\mathbb R^n)g,h∈Γ0​(Rn) satisfy the standing inclusions (3) and let x∗∈dom⁡hx^*\in\operatorname{dom}hx∗∈domh. Then ∂εh(x∗)⊂∂εg(x∗)\partial_\varepsilon h(x^*)\subset\partial_\varepsilon g(x^*)∂ε​h(x∗)⊂∂ε​g(x∗) for every ε>0\varepsilon>0ε>0 if and only if

∀ε>0, ∀y∈Y:⟨x∗,y⟩+ε≥h(x∗)+h∗(y) ⟹ ⟨x∗,y⟩+ε≥g(x∗)+g∗(y).(8)\forall\varepsilon>0,\ \forall y\in Y:\quad \langle x^*,y\rangle+\varepsilon\ge h(x^*)+h^*(y)\ \Longrightarrow\ \langle x^*,y\rangle+\varepsilon\ge g(x^*)+g^*(y).\tag{8}∀ε>0, ∀y∈Y:⟨x∗,y⟩+ε≥h(x∗)+h∗(y) ⟹ ⟨x∗,y⟩+ε≥g(x∗)+g∗(y).(8)

The equivalence unfolds the definition of the ε\varepsilonε-subdifferential through the conjugate: for x∗∈dom⁡θx^*\in\operatorname{dom}\thetax∗∈domθ, y∈∂εθ(x∗)y\in\partial_\varepsilon\theta(x^*)y∈∂ε​θ(x∗) exactly when θ(x∗)+θ∗(y)≤⟨x∗,y⟩+ε\theta(x^*)+\theta^*(y)\le\langle x^*,y\rangle+\varepsilonθ(x∗)+θ∗(y)≤⟨x∗,y⟩+ε. It is the second step of the proof of Theorem 3.1(i).

Formalization Note The sums h(x∗)+h∗(y)h(x^*)+h^*(y)h(x∗)+h∗(y) and g(x∗)+g∗(y)g(x^*)+g^*(y)g(x∗)+g∗(y) are extended reals and may be +∞+\infty+∞; the left-hand sides ⟨x∗,y⟩+ε\langle x^*,y\rangle+\varepsilon⟨x∗,y⟩+ε are real. The variable yyy, free in the paper's (8), is quantified over all of YYY. The standing assumptions are carried as hypotheses, though only g,h∈Γ0g,h\in\Gamma_0g,h∈Γ0​ is needed here.

Preamble
import Mathlib
import Definitions.Def_TaoAnDCA_GlobalOpt_Setting

open TaoAnDCA.GlobalOpt
Formal statement
namespace TaoAnDCA.GlobalOpt

theorem epsSubdiff_subset_iff_eq8 {n : ℕ} (g h : EuclideanSpace ℝ (Fin n) → EReal)
    (hgh : DCStanding g h) (x : EuclideanSpace ℝ (Fin n)) (hx : x ∈ effDom h) :
    (∀ ε : ℝ, 0 < ε → epsSubdiff h ε x ⊆ epsSubdiff g ε x) ↔
      ∀ ε : ℝ, 0 < ε → ∀ y : EuclideanSpace ℝ (Fin n),
        ((inner ℝ x y + ε : ℝ) : EReal) ≥ h x + CondatPD.FinDim.conj h y →
          ((inner ℝ x y + ε : ℝ) : EReal) ≥ g x + CondatPD.FinDim.conj g y := by sorry

end TaoAnDCA.GlobalOpt
Source
Pham Dinh & Le Thi, A d.c. optimization algorithm for solving the trust-region subproblem, SIAM J. Optim. 8 (1998), p. 484, §3.1, proof of Theorem 3.1, display (8)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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