Theorem 4.2, p. 12 — under Assumption 4.1, sup over the Wasserstein ball of E[max_k ℓ_k] equals the optimal value of the convex program (11)
ProvedWassDDRO.Reduction.theorem_4_2Let be a finite-dimensional real normed space (the paper's with an arbitrary norm ) with its Borel σ-algebra, and the dual norm. Let , let () be measurable with , let () be samples with empirical distribution , and let be the set of probability measures supported on within 1-Wasserstein distance of .
Theorem 4.2 (Convex reduction). If Assumption 4.1 holds ( convex and closed; each proper, convex and lower semicontinuous; no identically on ), then for any
where is the convex conjugate of and the support function of .
The theorem turns an optimization over infinitely many probability distributions into a finite-dimensional convex program, which is what makes Wasserstein distributionally robust optimization tractable.
Formalization Note The expectation of an extended-valued function uses the paper's convention , so a distribution with makes the supremum . The infimum of (11) is when (11) is infeasible. is the application of a continuous linear functional and its operator norm. The conditions , , and measurability of the are the paper's standing setting (§2, p. 5; p. 11).
import Mathlib import Definitions.Def_WassDDRO_Reduction_Setting open MeasureTheory Filter Topology
namespace WassDDRO.Reduction
/-- Theorem 4.2 (Convex reduction), p. 12. Under Assumption 4.1, for any `ε ≥ 0` the
worst-case expectation (10) equals the optimal value of the finite convex program (11). -/
theorem theorem_4_2 {E : Type*} [NormedAddCommGroup E] [NormedSpace ℝ E] [FiniteDimensional ℝ E]
[MeasurableSpace E] [BorelSpace E]
{K N : ℕ} (hK : 0 < K) (hN : 0 < N) (Ξ : Set E) (ℓ : Fin K → E → EReal)
(hmeas : ∀ k, Measurable (ℓ k)) (ξhat : Fin N → E) (hξ : ∀ i, ξhat i ∈ Ξ)
(ε : ℝ) (hε : 0 ≤ ε) (hA : Assumption41 Ξ ℓ) :
worstCaseExpectation ε Ξ ξhat (maxLoss ℓ) = program11Value ε Ξ ξhat ℓ := by sorry
end WassDDRO.Reduction
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.