Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 7 — f(xk)−Eikf(xk+1)≥38n(∥∇f(xk)∥1∗)2f(x_k)-\mathbb E_{i_k}f(x_{k+1})\ge\frac{3}{8n}(\|\nabla f(x_k)\|_1^*)^2f(xk​)−Eik​​f(xk+1​)≥8n3​(∥∇f(xk​)∥1∗​)2

Proved
NesterovRCD.Adaptive.expected_decrease

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

backtrackingconvex-optimizationcoordinate-descentp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1randomized-algorithms

Let n≥1n\ge1n≥1 and let f:Rn→Rf:\mathbb R^n\to\mathbb Rf:Rn→R be convex and differentiable with coordinate-wise Lipschitz partial derivatives (2.2) with constants L1,…,LnL_1,\dots,L_nL1​,…,Ln​. Write

∥g∥1∗=(∑i=1ngi2Li)1/2\|g\|_1^*=\Big(\sum_{i=1}^n\frac{g_i^2}{L_i}\Big)^{1/2}∥g∥1∗​=(i=1∑n​Li​gi2​​)1/2

for the dual norm of (2.7) with α=1\alpha=1α=1. Let x∈Rnx\in\mathbb R^nx∈Rn and let the estimates satisfy 0<L^i≤Li0<\hat L_i\le L_i0<L^i​≤Li​ for every iii. Draw the coordinate iii uniformly from {1,…,n}\{1,\dots,n\}{1,…,n} (the counter R0\mathcal R_0R0​) and let xi+x^+_ixi+​ be the point produced by one iteration of RACDM (6.1) on coordinate iii from the state (x,L^)(x,\hat L)(x,L^). Then

f(x)−∑i=1n1n f(xi+) ≥ 38n (∥∇f(x)∥1∗)2.f(x)-\sum_{i=1}^n\frac1n\,f(x^+_i)\ \ge\ \frac{3}{8n}\,\big(\|\nabla f(x)\|_1^*\big)^2 .f(x)−i=1∑n​n1​f(xi+​) ≥ 8n3​(∥∇f(x)∥1∗​)2.

This is the expected decrease of one RACDM iteration (p. 19), the analogue of (2.13) for the adaptive method; it is the inequality from which the rate (6.2) follows by the argument of Theorem 1.

Formalization Note Rn\mathbb R^nRn is EuclideanSpace ℝ (Fin n) (scalar coordinates: the paper sets N=nN=nN=n in §6.1), coordinates are indexed 0,…,n−10,\dots,n-10,…,n−1 for the paper's 1,…,n1,\dots,n1,…,n, and the gradient is an explicit map ggg with g(x)=∇f(x)g(x)=\nabla f(x)g(x)=∇f(x), as in the published definition ConvexOptAlg_CoordDescent_Defs that this mission reuses (its IsCoordSmooth f g L is (2.2) with one-dimensional blocks and ∣⋅∣|\cdot|∣⋅∣ as block norm; it includes differentiability). The uniform probabilities are the published pGamma L 0 i =Li0/∑jLj0=1/n=L_i^0/\sum_jL_j^0=1/n=Li0​/∑j​Lj0​=1/n. The norm is the published wnormDual L 1. The hypotheses 0<L^i≤Li0<\hat L_i\le L_i0<L^i​≤Li​ for all iii are what item 1 of Theorem 7 provides at the beginning of each iteration; n≥1n\ge1n≥1 is implicit in the paper (for n=0n=0n=0 there are no coordinates to draw).

Preamble
import Mathlib
import Definitions.Def_ConvexOptAlg_CoordDescent_Defs
import Definitions.Def_NesterovRCD_Adaptive_RACDM
Formal statement
namespace NesterovRCD.Adaptive

open ConvexOptAlg.CoordDescent

/-- Proof of Theorem 7, p. 19: for convex `f` and estimates `0 < L̂_i ≤ L_i`, one iteration of
RACDM (6.1) with the coordinate drawn uniformly (the counter `R_0`, `p_0(i) = 1/n`) satisfies
`f(x) − E_i f(x⁺) ≥ 3 (‖∇f(x)‖*_1)² / (8n)`. -/
theorem expected_decrease {n : ℕ} (hn : 0 < n) (f : Vec n → ℝ) (g : Vec n → Vec n)
    (hconv : ConvexOn ℝ Set.univ f) (L : Fin n → ℝ) (hL : IsCoordSmooth f g L)
    (x : Vec n) (Lh : Fin n → ℝ) (hLh : ∀ i, 0 < Lh i ∧ Lh i ≤ L i) :
    3 / (8 * n) * wnormDual L 1 (g x) ^ 2 ≤
      f x - ∑ i, pGamma L 0 i * f (racdmStep g (x, Lh) i).1 := by sorry

end NesterovRCD.Adaptive
Source
Nesterov, Efficiency of coordinate descent methods on huge-scale optimization problems, CORE Discussion Paper 2010/2, p. 19, proof of Theorem 7 (second display)
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