Theorem 1, Step 2 — every limit point is a zero
ProvedBregmanPPA.Convergence.step2bregman-functionmonotone-operatorp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1proximal-point
Let , , , , and satisfy the standing hypotheses of Theorem 1, and assume either (C1), , or (C2), for a proper lower semicontinuous convex function . Suppose has a zero. Every subsequential limit of is a zero of :
This identifies all possible accumulation points before whole-sequence convergence is established.
Formalization Note A subsequence has strictly increasing natural-number indices. The standing hypotheses include , with a positive uniform lower bound, and the inclusion form (4) of the run.
Preamble
import Mathlib import Definitions.Def_BregmanPPA_Convergence_Model open ThreeOpSplitting.Convergence InertialFB.IFB Filter Topology
Formal statement
namespace BregmanPPA.Convergence
variable {H : Type*} [NormedAddCommGroup H] [InnerProductSpace ℝ H] [FiniteDimensional ℝ H]
/-- Theorem 1, proof, Step 2: every subsequential limit is a zero. -/
theorem step2 (T : H → Set H) (S : Set H) (h : H → ℝ) (c : ℕ → ℝ) (x : ℕ → H)
(hT : IsMaximalMonotone T) (hh : IsBregmanFunction S h)
(hdom : dom T ⊆ closure S) (hc : ∀ k, 0 < c k)
(hcinf : ∃ ε : ℝ, 0 < ε ∧ ∀ k, ε ≤ c k)
(hx : IsBregmanPPARun S h T c x)
(hC : closure (dom T) ⊆ S ∨ ∃ f : H → EReal,
IsProperFn f ∧ IsConvexFn f ∧ LowerSemicontinuous f ∧ T = subdiffOp f)
(hz : (zer T).Nonempty) :
∀ (x' : H) (φ : ℕ → ℕ), StrictMono φ →
Tendsto (x ∘ φ) atTop (𝓝 x') → x' ∈ zer T := by sorry
end BregmanPPA.Convergence
Source
Eckstein, Nonlinear proximal point algorithms using Bregman functions, with applications to convex programming, Math. Oper. Res. 18(1) (1993), pp. 207–208, Theorem 1 proof, Step 2, https://doi.org/10.1287/moor.18.1.202
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.