Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§7.1.1: for g=λ∥⋅∥1g=\lambda\|\cdot\|_1g=λ∥⋅∥1​ the proximal step is soft thresholding Sλ/NρS_{\lambda/N\rho}Sλ/Nρ​

Proved
BoydADMM.Consensus.l1_prox_soft_threshold

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

admmconsensuslassop2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-bookp2o-v1soft-thresholding

Let N≥1N\ge1N≥1, ρ>0\rho>0ρ>0, λ>0\lambda>0λ>0 and g(z)=λ∥z∥1g(z)=\lambda\|z\|_1g(z)=λ∥z∥1​. With v=xˉk+1+(1/ρ)yˉkv=\bar x^{k+1}+(1/\rho)\bar y^kv=xˉk+1+(1/ρ)yˉ​k, the proximal step of the regularized consensus zzz-update,

minimize⁡z∈Rn λ∥z∥1+(Nρ/2) ∥z−v∥22,\operatorname*{minimize}_{z\in\mathbb R^n}\ \lambda\|z\|_1+(N\rho/2)\,\|z-v\|_2^2,z∈Rnminimize​ λ∥z∥1​+(Nρ/2)∥z−v∥22​,

has the unique solution

zj=Sλ/Nρ(vj),j=1,…,n,z_j=S_{\lambda/N\rho}(v_j),\qquad j=1,\dots,n,zj​=Sλ/Nρ​(vj​),j=1,…,n,

where SκS_\kappaSκ​ is soft thresholding.

Formalization Note The book prints zk+1:=Sλ/Nρ(xˉk+1−(1/ρ)yˉk)z^{k+1}:=S_{\lambda/N\rho}(\bar x^{k+1}-(1/\rho)\bar y^k)zk+1:=Sλ/Nρ​(xˉk+1−(1/ρ)yˉ​k); the intended reading, forced by the proximal form of the zzz-update displayed just above it on the same page, is xˉk+1+(1/ρ)yˉk\bar x^{k+1}+(1/\rho)\bar y^kxˉk+1+(1/ρ)yˉ​k, which is what is stated. λ\lambdaλ is named lam in Lean. The statement is an "iff" (existence and uniqueness of the minimizer).

Preamble
import Mathlib
import Definitions.Def_BoydADMM_Consensus_Model

open scoped BigOperators InnerProductSpace
Formal statement
namespace BoydADMM.Consensus

/-- §7.1.1, p. 52 (sign corrected): for `g(z) = λ‖z‖₁` with `λ > 0` (`lam` in Lean), the
proximal step `argmin_z (λ‖z‖₁ + (Nρ/2)‖z − x̄^{k+1} − (1/ρ)ȳ^k‖²)` has the unique solution
`z = S_{λ/Nρ}(x̄^{k+1} + (1/ρ)ȳ^k)`, componentwise. The book prints `x̄^{k+1} − (1/ρ)ȳ^k`. -/
theorem l1_prox_soft_threshold {n N : ℕ} (hN : 0 < N) (ρ : ℝ) (hρ : 0 < ρ)
    (lam : ℝ) (hlam : 0 < lam)
    (x y : Fin N → EuclideanSpace ℝ (Fin n)) (z : EuclideanSpace ℝ (Fin n)) :
    (∀ z' : EuclideanSpace ℝ (Fin n),
        lam * ∑ j, |z j| + ((N : ℝ) * ρ / 2) * ‖z - avg x - (1 / ρ) • avg y‖ ^ 2 ≤
          lam * ∑ j, |z' j| + ((N : ℝ) * ρ / 2) * ‖z' - avg x - (1 / ρ) • avg y‖ ^ 2) ↔
      ∀ j, z j = BoydADMM.Prox.softThreshold (lam / ((N : ℝ) * ρ)) ((avg x + (1 / ρ) • avg y) j) := by sorry

end BoydADMM.Consensus
Source
Boyd, Parikh, Chu, Peleato, Eckstein, Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers, Found. Trends Mach. Learn. 3(1) (2011), p. 52, §7.1.1 (example g(z) = λ‖z‖₁; the book prints x̄^{k+1} − (1/ρ)ȳ^k, corrected to +)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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