Proof of Theorem 4.2, p. 13 — under Assumption 4.1, the optimal values of (12f) and (11) coincide
ProvedWassDDRO.Reduction.program12f_eq_program11convex-conjugateinf-convolutionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1support-functionwasserstein
Let , , and samples be as in Theorem 4.2 (), and suppose Assumption 4.1 holds. Then for every the optimal value of (12f),
equals the optimal value of (11),
The constraint of (11) replaces the conjugate of the sum by the inf-convolution of the conjugates of and (the support function ). The two constraint functions differ by a lower-semicontinuous closure, so the statement is an equality of optimal values, not of feasible sets.
Formalization Note The sum is taken in EReal. Under Assumption 4.1 neither summand is ( is finite somewhere and contains the samples), so Mathlib's never arises.
Preamble
import Mathlib import Definitions.Def_WassDDRO_Reduction_Setting open MeasureTheory Filter Topology
Formal statement
namespace WassDDRO.Reduction
/-- Proof of Theorem 4.2, p. 13: under Assumption 4.1, for any `ε ≥ 0` the optimal values of
(12f) and (11) coincide. -/
theorem program12f_eq_program11 {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) :
program12fValue ε Ξ ξhat ℓ = program11Value ε Ξ ξhat ℓ := by sorry
end WassDDRO.Reduction
Source
Mohajerin Esfahani & Kuhn, arXiv:1505.05116v3, proof of Theorem 4.2, p. 13 ("(12f) is indeed equivalent to (11) under Assumption 4.1"); program (11), p. 12
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.