Proposition 3.3 — 𝒞 is nonempty and closed, F is locally Lipschitz, and the level sets of F on 𝒞 are compact
ProvedProgHedging.Convex.proposition_3_3p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1progressive-hedgingstochastic-programming
In problem (P), the admissible set is nonempty and closed, the objective is locally Lipschitz continuous on the policy space , and every level set
is compact.
This transfers the standing assumptions on the scenario data to the full problem, and yields existence of optimal solutions of (P) whenever it is feasible.
Formalization Note. The printed statement breaks off after "all level sets of the form (3.8)" without a predicate; its proof ends "It follows that any set (3.8) is compact", which is the reading formalized. Local Lipschitz continuity and compactness use Mathlib's (product) topology on policies, which in finite dimension is the topology of the paper's norm.
Preamble
import Mathlib import Definitions.Def_ProgHedging_Convex_Problem open scoped RealInnerProductSpace Pointwise Topology open Filter
Formal statement
namespace ProgHedging.Convex
/-- Proposition 3.3, p. 13 (with the conclusion of its proof: the level sets are compact).
The feasible set `𝒞` of (P) is nonempty and closed, `F` is locally Lipschitz on `ℰ`, and every level
set `{X ∈ 𝒞 | F(X) ≤ α}` is compact. -/
theorem proposition_3_3 {S : Type*} [Fintype S] {n T : ℕ} (pr : Problem S n T) :
pr.adm.Nonempty ∧ IsClosed pr.adm ∧ LocallyLipschitz pr.F ∧
∀ α : ℝ, IsCompact {X | X ∈ pr.adm ∧ pr.F X ≤ α} := by sorry
end ProgHedging.Convex
Source
Rockafellar and Wets, Scenarios and policy aggregation in optimization under uncertainty, IIASA Working Paper WP-87-119 (1987), p. 13, Proposition 3.3, (3.8), and its proof
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.