Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(2.13) — f(x)−∑ipα(i)f(Ti(x))≥12Sα(∥∇f(x)∥1−α∗)2f(x)-\sum_ip_\alpha^{(i)}f(T_i(x))\ge\frac1{2S_\alpha}(\|\nabla f(x)\|^*_{1-\alpha})^2f(x)−∑i​pα(i)​f(Ti​(x))≥2Sα​1​(∥∇f(x)∥1−α∗​)2

Proved
NesterovRCD.Sublinear.expected_decrease

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

coordinate-descentp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1randomized-algorithms

Let n≥1n\ge1n≥1, let fff satisfy (2.2) with constants Li>0L_i>0Li​>0 on finite-dimensional normed blocks, let α∈R\alpha\in\mathbb Rα∈R, and let pα(i)=Liα/Sαp_\alpha^{(i)}=L_i^\alpha/S_\alphapα(i)​=Liα​/Sα​ be the probabilities (2.5). For an arbitrary choice of the vectors s#s^\#s#, every point xxx satisfies

f(x)−∑i=1npα(i)f(Ti(x))≥12Sα(∥∇f(x)∥1−α∗)2.f(x)-\sum_{i=1}^np_\alpha^{(i)}f(T_i(x))\ge\frac1{2S_\alpha}\big(\|\nabla f(x)\|^*_{1-\alpha}\big)^2 .f(x)−i=1∑n​pα(i)​f(Ti​(x))≥2Sα​1​(∥∇f(x)∥1−α∗​)2.

The left side is the expected decrease f(xk)−Eikf(xk+1)f(x_k)-E_{i_k}f(x_{k+1})f(xk​)−Eik​​f(xk+1​) of one iteration of RCDM(α,x0)(\alpha,x_0)(α,x0​) at xk=xx_k=xxk​=x; the right side expresses it through the full gradient in the dual weighted norm.

Formalization Note The statement is pointwise in xxx, which is how the paper uses it for the random iterate xkx_kxk​. n≥1n\ge1n≥1 is explicit (for n=0n=0n=0 the probabilities are 0/00/00/0). Convexity is not needed.

Preamble
import Mathlib
import Definitions.Def_NesterovRCD_Sublinear_Basic
Formal statement
namespace NesterovRCD.Sublinear

variable {n : ℕ} {E : Fin n → Type*} [∀ i, NormedAddCommGroup (E i)] [∀ i, NormedSpace ℝ (E i)]
  [∀ i, FiniteDimensional ℝ (E i)]

theorem expected_decrease (hn : 0 < n) (f : Blocks E → ℝ) (L : Fin n → ℝ) (hL : CoordLipschitz f L)
    (sharp : ∀ i, (E i →L[ℝ] ℝ) → E i) (hsharp : IsSharpSelection sharp)
    (α : ℝ) (x : Blocks E) :
    wdual L (1 - α) (gradBlocks f x) ^ 2 / (2 * S L α)
      ≤ f x - ∑ i, prob L α i * f (coordStep f L sharp i x) := by sorry

end NesterovRCD.Sublinear
Source
Nesterov, Efficiency of coordinate descent methods on huge-scale optimization problems, CORE Discussion Paper 2010/2, p. 7, proof of Theorem 1, (2.13)
Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 7, 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