§2.3, p. 131 — proximal minimizers satisfy the subdifferential optimality condition
ProvedProxAlg.FixedPoint.prox_optimalityoptimality-conditionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1proximal-operatorsubdifferential
Let be closed, proper and convex, and let . Write for the subdifferential defined by (2.3). Then
equivalently, .
This characterizes the proximal minimizer through a first-order condition and supplies the link between proximal maps and subdifferentials.
Formalization Note The proximal point is expressed by Moreau's IsProx predicate, which minimizes and has a unique solution under the stated hypotheses. The paper's subdifferential is the local definition from (2.3).
Preamble
import Mathlib import Definitions.Def_MoreauProx_Decomposition_ConvexDuality import Definitions.Def_ProxAlg_FixedPoint_Basic
Formal statement
namespace ProxAlg.FixedPoint
theorem prox_optimality {n : ℕ} (f : EuclideanSpace ℝ (Fin n) → EReal)
(hf : MoreauProx.Decomposition.GammaZero f)
(v x : EuclideanSpace ℝ (Fin n)) :
MoreauProx.Decomposition.IsProx f v x ↔ v - x ∈ subdifferential f x := by sorry
end ProxAlg.FixedPoint
Source
Parikh & Boyd, Proximal Algorithms, Found. Trends Optim. 1(3) (2014), §2.3, p. 131, proof following the fixed-point statement
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.