(15), p. 11 — at a limit point z of the BSUM iterates, lim f(xʳ) = f(z)
ProvedBSUMConv.BSUM.eq_15block-coordinate-descentconvergencep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1successive-upper-bound-minimization
In the setting of Theorem 2 (closed convex blocks , continuous on , Assumption 2, BSUM iterates under the cyclic rule), let be a limit point of the sequence . Then
Together with (14) this identifies the limit of the objective values, which the rest of the proof of Theorem 2(a) uses repeatedly.
Formalization Note "Limit point" is MapClusterPt z atTop x.
Preamble
import Mathlib import Definitions.Def_TsengBCD_Stationary_Setting import Definitions.Def_BSUMConv_BSUM_Setting
Formal statement
namespace BSUMConv.BSUM
open TsengBCD.Stationary Filter Topology
/-- (15), p. 11: if `z` is a limit point of a cyclic BSUM run, then
`lim_{r→∞} f(x^r) = f(z)`. -/
theorem eq_15 {N : ℕ} {n : Fin N → ℕ}
(Xs : (i : Fin N) → Set (EuclideanSpace ℝ (Fin (n i))))
(hXconv : ∀ i, Convex ℝ (Xs i)) (hXclosed : ∀ i, IsClosed (Xs i))
(f : X n → ℝ) (hf : ContinuousOn f (Xset Xs))
(u : (i : Fin N) → EuclideanSpace ℝ (Fin (n i)) → X n → ℝ) (hA2 : Assumption2 Xs f u)
(s : ℕ → Fin N) (hs : IsCyclic s) (x : ℕ → X n) (hx : IsBSUMRun Xs u s x)
(z : X n) (hz : MapClusterPt z atTop x) :
Tendsto (fun r => f (x r)) atTop (𝓝 (f z)) := by sorry
end BSUMConv.BSUM
Source
Razaviyayn, Hong & Luo, arXiv:1209.2385v1, p. 11, proof of Theorem 2(a), display (15)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.