Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 6 — the nonquadratic method of multipliers (9) finds an optimal multiplier, and limit points of xkx^kxk solve (7)

Proved
BregmanPPA.EqMult.theorem6

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

bregman-functionconvex-optimizationlagrangian-dualitymethod-of-multipliersp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Consider the convex program

minimize f(x)such thatAx=b, x∈X,(7)\text{minimize } f(x)\quad\text{such that}\quad Ax = b,\ x \in X, \tag{7}minimize f(x)such thatAx=b, x∈X,(7)

where f:Rn→(−∞,+∞]f : \mathbb R^n \to (-\infty,+\infty]f:Rn→(−∞,+∞] is lower semicontinuous and convex, AAA is an m×nm\times nm×n matrix, b∈Rmb\in\mathbb R^mb∈Rm, X⊆RnX\subseteq\mathbb R^nX⊆Rn is closed and convex, and inf⁡{f(x)∣x∈X}<∞\inf\{f(x)\mid x\in X\} < \inftyinf{f(x)∣x∈X}<∞. Its dual functional is

d(p)=inf⁡x∈X{f(x)+⟨p,Ax−b⟩},d(p) = \inf_{x\in X}\bigl\{ f(x) + \langle p, Ax - b\rangle \bigr\},d(p)=x∈Xinf​{f(x)+⟨p,Ax−b⟩},

assumed not identically −∞-\infty−∞; an optimal (Lagrange) multiplier is a maximizer of ddd over Rm\mathbb R^mRm.

Let hhh be a Bregman function with zone Rm\mathbb R^mRm and im⁡(∇h)=Rm\operatorname{im}(\nabla h) = \mathbb R^mim(∇h)=Rm, let h∗h^*h∗ be its convex conjugate, and let {ck}k≥0\{c_k\}_{k\ge0}{ck​}k≥0​ be positive scalars with positive infimum. Let {(xk,pk)}k≥0⊆Rn×Rm\{(x^k,p^k)\}_{k\ge0} \subseteq \mathbb R^n\times\mathbb R^m{(xk,pk)}k≥0​⊆Rn×Rm conform to the recursion

xk+1∈arg min⁡x∈X{f(x)+1ckh∗(∇h(pk)+ck(Ax−b))},pk+1=∇h∗(∇h(pk)+ck(Axk+1−b)).(9)x^{k+1} \in \operatorname*{arg\,min}_{x\in X}\Bigl\{ f(x) + \tfrac{1}{c_k} h^*\bigl(\nabla h(p^k) + c_k(Ax - b)\bigr)\Bigr\},\qquad p^{k+1} = \nabla h^*\bigl(\nabla h(p^k) + c_k(Ax^{k+1} - b)\bigr). \tag{9}xk+1∈x∈Xargmin​{f(x)+ck​1​h∗(∇h(pk)+ck​(Ax−b))},pk+1=∇h∗(∇h(pk)+ck​(Axk+1−b)).(9)

Then:

  1. if an optimal multiplier exists, {pk}\{p^k\}{pk} converges to an optimal multiplier, and every limit point of {xk}\{x^k\}{xk} solves (7);
  2. if no optimal multiplier exists, {pk}\{p^k\}{pk} is unbounded.

With h(p)=12∥p∥2h(p) = \frac12\|p\|^2h(p)=21​∥p∥2, (9) is the classical (quadratic) method of multipliers; the theorem covers the whole family of nonquadratic penalties generated by Bregman functions with zone Rm\mathbb R^mRm and surjective gradient.

Formalization Note AAA is a continuous linear map between EuclideanSpace ℝ (Fin n) and EuclideanSpace ℝ (Fin m). "Zone Rm\mathbb R^mRm" is the Bregman structure with SSS the whole space; im⁡∇h=Rm\operatorname{im}\nabla h = \mathbb R^mim∇h=Rm is surjectivity of ∇h\nabla h∇h. Recursion (9) uses the real value of h∗h^*h∗, which is finite everywhere under im⁡∇h=Rm\operatorname{im}\nabla h = \mathbb R^mim∇h=Rm; the arg min is a minimality predicate. "Limit point" is a subsequential limit; "unbounded" is "the range of k↦pkk \mapsto p^kk↦pk is not bounded". This statement is the first two sentences of the conclusion of Theorem 6; the third (existence of infinite sequences) is the separate item theorem6_existence.

Preamble
import Mathlib
import Definitions.Def_InertialFB_IFB_ConvexAnalysis
import Definitions.Def_fenchelConjugate
import Definitions.Def_BregmanPPA_Convergence_Model
import Definitions.Def_BregmanPPA_EqMult_EqConstrainedProgram

open InertialFB.IFB ConvexOptimization Filter Topology
Formal statement
namespace BregmanPPA.EqMult
theorem theorem6 {n m : ℕ}
    (A : EuclideanSpace ℝ (Fin n) →L[ℝ] EuclideanSpace ℝ (Fin m))
    (b : EuclideanSpace ℝ (Fin m)) (X : Set (EuclideanSpace ℝ (Fin n)))
    (f : EuclideanSpace ℝ (Fin n) → EReal)
    (hP : IsEqProgram X f) (hd : ∃ p, dualEq A b X f p ≠ ⊥)
    (h : EuclideanSpace ℝ (Fin m) → ℝ) (hh : BregmanPPA.Convergence.IsBregmanFunction Set.univ h)
    (him : Function.Surjective (gradient h))
    (c : ℕ → ℝ) (hc : ∀ k, 0 < c k) (hcinf : ∃ ε > 0, ∀ k, ε ≤ c k)
    (x : ℕ → EuclideanSpace ℝ (Fin n)) (p : ℕ → EuclideanSpace ℝ (Fin m))
    (hrun : IsEqMultRun A b X f h c x p) :
    ((∃ q, IsOptimalMultiplier A b X f q) →
      (∃ pstar, IsOptimalMultiplier A b X f pstar ∧ Tendsto p atTop (𝓝 pstar)) ∧
      ∀ (φ : ℕ → ℕ) (x' : EuclideanSpace ℝ (Fin n)), StrictMono φ →
        Tendsto (x ∘ φ) atTop (𝓝 x') → IsSolution7 A b X f x') ∧
    ((¬ ∃ q, IsOptimalMultiplier A b X f q) → ¬ Bornology.IsBounded (Set.range p)) := by sorry
end BregmanPPA.EqMult
Source
Eckstein, Nonlinear proximal point algorithms using Bregman functions, with applications to convex programming, Math. Oper. Res. 18(1) (1993), pp. 212–213, Theorem 6 (first two sentences of the conclusion)
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