Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 1 — the Bregman proximal point algorithm converges to a zero of a maximal monotone T

Proved
BregmanPPA.ProxMult.theorem1

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

bregman-functionmaximal-monotonep2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1proximal-point

Let HHH be a finite-dimensional real inner product space, T:H⇉HT:H\rightrightarrows HT:H⇉H a maximal monotone operator, and hhh a Bregman function with zone SSS, S‾⊇dom⁡T\overline S\supseteq\operatorname{dom}TS⊇domT. Let {ck}\{c_k\}{ck​} be positive scalars with inf⁡kck>0\inf_k c_k>0infk​ck​>0, and let {xk}\{x^k\}{xk} be an infinite sequence conforming to

xk+1=(∇h+ckT)−1(∇h(xk)).(3)x^{k+1}=(\nabla h+c_kT)^{-1}\big(\nabla h(x^k)\big).\tag{3}xk+1=(∇h+ck​T)−1(∇h(xk)).(3)

Suppose that (C1) S⊇dom⁡T‾S\supseteq\overline{\operatorname{dom}T}S⊇domT, or (C2) T=∂fT=\partial fT=∂f for a proper lower semicontinuous convex f:H→(−∞,+∞]f:H\to(-\infty,+\infty]f:H→(−∞,+∞]. Then if TTT has a zero, {xk}\{x^k\}{xk} converges to one of them. If TTT has no zero and (C1) holds, {xk}\{x^k\}{xk} is unbounded.

This is the paper's central theorem, restated in this mission so that it can be applied to the saddle operator KKK on Rn+m\mathbb R^{n+m}Rn+m.

Formalization Note The paper states the theorem on Rn\mathbb R^nRn (and prints "maximal monontone operator of Rn\mathbb R^nRn"); the restatement is over any finite-dimensional real inner product space, which is Rn\mathbb R^nRn up to isometry and includes Rn+m\mathbb R^{n+m}Rn+m as WithLp 2 (E n × E m). (3) is encoded by (4) with xk∈Sx^k\in Sxk∈S. "Converges to one of them" is ∃z, 0∈Tz∧xk→z\exists z,\ 0\in T z\wedge x^k\to z∃z, 0∈Tz∧xk→z; "unbounded" is "the range of xxx is not bounded".

Preamble
import Mathlib
import Definitions.Def_BregmanPPA_Convergence_Model

open Filter Topology ThreeOpSplitting.Convergence InertialFB.IFB
Formal statement
namespace BregmanPPA.ProxMult

variable {H : Type*} [NormedAddCommGroup H] [InnerProductSpace ℝ H] [FiniteDimensional ℝ H]

/-- Theorem 1, pp. 206–207, restated on a finite-dimensional inner product space `H` (it is
applied on `ℝⁿ⁺ᵐ`): convergence to a zero, or unboundedness when there is none under (C1). -/
theorem theorem1 (T : H → Set H) (S : Set H) (h : H → ℝ) (c : ℕ → ℝ) (x : ℕ → H)
    (hT : IsMaximalMonotone T) (hh : BregmanPPA.Convergence.IsBregmanFunction S h)
    (hdom : dom T ⊆ closure S) (hc : ∀ k, 0 < c k)
    (hcinf : ∃ ε : ℝ, 0 < ε ∧ ∀ k, ε ≤ c k)
    (hx : BregmanPPA.Convergence.IsBregmanPPARun S h T c x)
    (hC : closure (dom T) ⊆ S ∨ ∃ f : H → EReal,
      IsProperFn f ∧ IsConvexFn f ∧ LowerSemicontinuous f ∧ T = BregmanPPA.Convergence.subdiffOp f) :
    ((zer T).Nonempty → ∃ z : H, z ∈ zer T ∧ Tendsto x atTop (𝓝 z)) ∧
    (zer T = ∅ → closure (dom T) ⊆ S → ¬ Bornology.IsBounded (Set.range x)) := by sorry

end BregmanPPA.ProxMult
Source
Eckstein, Nonlinear proximal point algorithms using Bregman functions, with applications to convex programming, Math. Oper. Res. 18(1) (1993), pp. 206–207, Theorem 1
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