Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 4, p. 13 — E_{k+1} − E_k + h(h/2·C_k² + δ_kβ_{k+1})‖∇f(x_{k+1})‖² ≤ 0

Proved
HessianDamping.IPAHD.EK_succ_sub_add_le

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

hessian-dampinginertial-algorithmlyapunovp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Let H\mathcal HH be a real Hilbert space and f:H→Rf:\mathcal H\to\mathbb Rf:H→R a convex C1\mathcal C^1C1 function with a minimizer x⋆x^\starx⋆. Let α≥1\alpha\ge1α≥1, h>0h>0h>0, s=h2s=h^2s=h2, let (βk)(\beta_k)(βk​), (bk)(b_k)(bk​) be non-negative sequences, and let (xk)(x_k)(xk​) be generated by (IPAHD). If k≥1k\ge1k≥1 and both growth conditions hold at kkk, bkhk−βk+1−k(βk+1−βk)>0b_khk-\beta_{k+1}-k(\beta_{k+1}-\beta_k)>0bk​hk−βk+1​−k(βk+1​−βk​)>0 and δk+1−δk≤(α−1)δkk+1\delta_{k+1}-\delta_k\le(\alpha-1)\frac{\delta_k}{k+1}δk+1​−δk​≤(α−1)k+1δk​​, then

Ek+1−Ek+h(h2(βk+1+k(βk+1−βk)−bkhk)2+δkβk+1)∥∇f(xk+1)∥2≤0,E_{k+1}-E_k+h\Big(\frac h2\big(\beta_{k+1}+k(\beta_{k+1}-\beta_k)-b_khk\big)^2+\delta_k\beta_{k+1}\Big)\|\nabla f(x_{k+1})\|^2\le0,Ek+1​−Ek​+h(2h​(βk+1​+k(βk+1​−βk​)−bk​hk)2+δk​βk+1​)∥∇f(xk+1​)∥2≤0,

where δk\delta_kδk​ is (9) and EkE_kEk​ the energy of the proof of Theorem 4.

Summing these inequalities over kkk gives Theorem 4 (ii).

Formalization Note. The page sums the inequalities over all kkk for which the growth conditions hold; here the per-step inequality is stated at a single k≥1k\ge1k≥1 with the growth conditions at that kkk. x⋆x^\starx⋆ is a point with f(x⋆)≤f(y)f(x^\star)\le f(y)f(x⋆)≤f(y) for all yyy; h>0h>0h>0, βk,bk≥0\beta_k,b_k\ge0βk​,bk​≥0 are implicit on the page and added.

Preamble
import Mathlib
import Definitions.Def_GoldenRatioVI_Shared_IsProxPoint
import Definitions.Def_HessianDamping_IPAHD_Setting
Formal statement
namespace HessianDamping.IPAHD

/-- Proof of Theorem 4, p. 13: the inequality that is summed to obtain (ii), under (G₂dis)
and (G₃dis) at `k`. -/
theorem EK_succ_sub_add_le
    {H : Type*} [NormedAddCommGroup H] [InnerProductSpace ℝ H] [CompleteSpace H]
    (f : H → ℝ) (hfconv : ConvexOn ℝ Set.univ f) (hfC1 : ContDiff ℝ 1 f)
    (xstar : H) (hxstar : ∀ y, f xstar ≤ f y)
    (α : ℝ) (hα : 1 ≤ α)
    (h s : ℝ) (hh : 0 < h) (hs : s = h ^ 2)
    (β b : ℕ → ℝ) (hβ : ∀ k, 0 ≤ β k) (hb : ∀ k, 0 ≤ b k)
    (x : ℕ → H) (hx : IsIPAHDRun f α s β b x)
    (k : ℕ) (hk : 1 ≤ k)
    (hG2 : 0 < b k * h * k - β (k + 1) - k * (β (k + 1) - β k))
    (hG3 : deltaK h β b (k + 1) - deltaK h β b k ≤ (α - 1) * deltaK h β b k / ((k : ℝ) + 1)) :
    EK f α h β b x xstar (k + 1) - EK f α h β b x xstar k
      + h * (h / 2 * CK h β b k ^ 2 + deltaK h β b k * β (k + 1))
        * ‖gradient f (x (k + 1))‖ ^ 2 ≤ 0 := by sorry

end HessianDamping.IPAHD
Source
Attouch, Chbani, Fadili, Riahi, First-order optimization algorithms via inertial systems with Hessian driven damping, arXiv:1907.10536v2, §3.1.1, proof of Theorem 4, p. 13
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 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