Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(3.34) — the combined stochastic prox inequality under α_kλ_k ≤ β_k

Proved
NonconvexAG.StochComposite.eq_3_34

by mikedeng1 · Oct 9, 2026 · Mathlib 0df444a (Lean v4.33.1)

accelerated-gradientcomposite-optimizationp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

In the setting of (3.32)–(3.33), for every k≥1k\ge1k≥1 with αkλk≤βk\alpha_k\lambda_k\le\beta_kαk​λk​≤βk​ and every x∈Kx\in Kx∈K,

⟨∇Ψ(xkmd)+δˉk,xkag−αkx−(1−αk)xk−1ag⟩+X(xkag)≤(1−αk)X(xk−1ag)+αkX(x)+αk2λk[∥xk−1−x∥2−∥xk−x∥2]−12βk∥xkag−xkmd∥2.\langle\nabla\Psi(x^{md}_k)+\bar\delta_k,x^{ag}_k-\alpha_kx-(1-\alpha_k)x^{ag}_{k-1}\rangle+\mathcal X(x^{ag}_k)\le(1-\alpha_k)\mathcal X(x^{ag}_{k-1})+\alpha_k\mathcal X(x)+\frac{\alpha_k}{2\lambda_k}\Big[\|x_{k-1}-x\|^2-\|x_k-x\|^2\Big]-\frac1{2\beta_k}\|x^{ag}_k-x^{md}_k\|^2 .⟨∇Ψ(xkmd​)+δˉk​,xkag​−αk​x−(1−αk​)xk−1ag​⟩+X(xkag​)≤(1−αk​)X(xk−1ag​)+αk​X(x)+2λk​αk​​[∥xk−1​−x∥2−∥xk​−x∥2]−2βk​1​∥xkag​−xkmd​∥2.

It combines the two three-point inequalities through the convexity of X\mathcal XX and the definition (2.2) of xkmdx^{md}_kxkmd​.

Formalization Note Pathwise, for x∈Kx\in Kx∈K. At k=1k=1k=1 the term X(x0ag)=X(x0)\mathcal X(x^{ag}_0)=\mathcal X(x_0)X(x0ag​)=X(x0​) carries the factor 1−α1=01-\alpha_1=01−α1​=0, so x0∈Kx_0\in Kx0​∈K is not needed (the Lean X is a real-valued function on all of Rn\mathbb R^nRn whose values off KKK are irrelevant).

Preamble
import Mathlib
import Definitions.Def_ConvexOptAlg_SmoothGD_Defs
import Definitions.Def_GhadimiLan_RSG_Model
import Definitions.Def_NonconvexAG_StochComposite_ProxMap
import Definitions.Def_NonconvexAG_StochComposite_RSAG
open MeasureTheory ProbabilityTheory
open scoped InnerProductSpace
Formal statement
namespace NonconvexAG.StochComposite

open GhadimiLan.RSG (E)
open ConvexOptAlg.SmoothGD

/-- (3.34), p. 22 (outer ends of the chain), along Algorithm 4, for every realization `ω`, every
`k ≥ 1` with `αₖλₖ ≤ βₖ` and every `x` in the domain `K` of `𝒳`. -/
theorem eq_3_34 {n : ℕ} (gΨ : E n → E n) (K : Set (E n)) (X : E n → ℝ)
    (hX : ConvexOn ℝ K X) (P : E n → E n → ℝ → E n) (hP : IsProxMap K X P)
    {Ξ : Type*} (G : E n → Ξ → E n)
    (α β lam : ℕ → ℝ) (m : ℕ → ℕ) (hstep : RSAGStepsizes α β lam m) (x0 : E n)
    {Ω : Type*} (ξ : ℕ → Ω → Ξ) :
    let xk : ℕ → Ω → E n := xSeq G P α β lam m x0 ξ
    let xag : ℕ → Ω → E n := xagSeq G P α β lam m x0 ξ
    let xmd : ℕ → Ω → E n := xmdSeq G P α β lam m x0 ξ
    let δ : ℕ → Ω → E n := deltaBar G gΨ P α β lam m x0 ξ
    ∀ ω : Ω, ∀ k, 1 ≤ k → α k * lam k ≤ β k → ∀ x ∈ K,
      inner ℝ (gΨ (xmd k ω) + δ k ω) (xag k ω - α k • x - (1 - α k) • xag (k - 1) ω) +
          X (xag k ω) ≤
        (1 - α k) * X (xag (k - 1) ω) + α k * X x +
          α k / (2 * lam k) * (‖xk (k - 1) ω - x‖ ^ 2 - ‖xk k ω - x‖ ^ 2) -
          1 / (2 * β k) * ‖xag k ω - xmd k ω‖ ^ 2 := by sorry

end NonconvexAG.StochComposite
Source
Ghadimi & Lan, Accelerated Gradient Methods for Nonconvex Nonlinear and Stochastic Programming, arXiv:1310.3787v1, p. 22, (3.34)
Human review
  • Endorsed by Shuze Chen · Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 9, 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