Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 4.1, fifth display, p. 17 — the gradient term split by the first line of (72)

Proved
NAGFlow.Implicit.gradient_split

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

convex-optimizationimplicit-schemelyapunov-functionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let VVV be a real Hilbert space, ∇f:V→V\nabla f:V\to V∇f:V→V any map, x∗∈Vx^*\in Vx∗∈V, (xk),(vk)(x_k),(v_k)(xk​),(vk​) sequences in VVV and (αk)(\alpha_k)(αk​) a real sequence. Fix kkk with αk>0\alpha_k>0αk​>0 and suppose the first line of the implicit scheme (72) holds at step kkk:

xk+1−xkαk=vk+1−xk+1.\frac{x_{k+1}-x_k}{\alpha_k}=v_{k+1}-x_{k+1}.αk​xk+1​−xk​​=vk+1​−xk+1​.

Then, writing vk+1−x∗=(vk+1−xk+1)+(xk+1−x∗)v_{k+1}-x^*=(v_{k+1}-x_{k+1})+(x_{k+1}-x^*)vk+1​−x∗=(vk+1​−xk+1​)+(xk+1​−x∗),

−αk⟨∇f(xk+1),vk+1−x∗⟩=−⟨∇f(xk+1),xk+1−xk⟩−αk⟨∇f(xk+1),xk+1−x∗⟩.-\alpha_k\langle\nabla f(x_{k+1}),v_{k+1}-x^*\rangle=-\langle\nabla f(x_{k+1}),x_{k+1}-x_k\rangle-\alpha_k\langle\nabla f(x_{k+1}),x_{k+1}-x^*\rangle .−αk​⟨∇f(xk+1​),vk+1​−x∗⟩=−⟨∇f(xk+1​),xk+1​−xk​⟩−αk​⟨∇f(xk+1​),xk+1​−x∗⟩.

The two terms on the right are the ones the μ\muμ-convexity inequality (2) bounds, at the pairs (xk,xk+1)(x_k,x_{k+1})(xk​,xk+1​) and (x∗,xk+1)(x^*,x_{k+1})(x∗,xk+1​).

Formalization Note. The duality pairing is the real inner product (Riesz). No property of ∇f\nabla f∇f is needed.

Preamble
import Mathlib
import Definitions.Def_NAGFlow_Implicit_Setting
Formal statement
namespace NAGFlow.Implicit

open scoped RealInnerProductSpace

/-- Proof of Theorem 4.1, fifth display, p. 17: if `α_k > 0` and the first line of (72) holds at
step `k`, then
`-α_k ⟪∇f(x_{k+1}), v_{k+1} - x*⟫
  = -⟪∇f(x_{k+1}), x_{k+1} - x_k⟫ - α_k ⟪∇f(x_{k+1}), x_{k+1} - x*⟫`. -/
theorem gradient_split {V : Type*} [NormedAddCommGroup V] [InnerProductSpace ℝ V]
    [CompleteSpace V] (gradf : V → V) (xstar : V) (α : ℕ → ℝ) (x v : ℕ → V) (k : ℕ)
    (hα : 0 < α k) (hx : (1 / α k) • (x (k + 1) - x k) = v (k + 1) - x (k + 1)) :
    -α k * ⟪gradf (x (k + 1)), v (k + 1) - xstar⟫
      = -⟪gradf (x (k + 1)), x (k + 1) - x k⟫
        - α k * ⟪gradf (x (k + 1)), x (k + 1) - xstar⟫ := by sorry

end NAGFlow.Implicit
Source
Luo & Chen, arXiv:1909.03145v4, proof of Theorem 4.1, fifth display ('For the gradient term …'), p. 17
Human review
  • Endorsed by Shuze Chen · Oct 10, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 10, 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