Theorem 1, p. 909 (upper-bound part) — sup over 𝔽₁ of E_ℙ((r⁰ + r′ζ̃)⁺) ≤ π¹(r⁰, r)
ProvedTractableDRO.Unified.theorem_1_upperdistributionally-robust-optimizationdualitymoment-boundsp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
Let be arbitrary sets and let be the family of probability distributions of a random vector with finite mean and . Then for every and ,
This is the upper-bound half of Theorem 1 of Goh and Sim; the paper also proves that the bound is tight under its standing assumptions. Theorem 4 uses only this half.
Formalization Note Both sides are extended reals: the left side is when is empty, and infinite suprema are allowed on the right. No convexity or closedness of is assumed; the upper bound does not need it.
Preamble
import Mathlib import Definitions.Def_TractableDRO_Unified_Model open MeasureTheory ProbabilityTheory Matrix
Formal statement
namespace TractableDRO.Unified
/-- Theorem 1, p. 909 (upper-bound part): `sup_{ℙ∈𝔽₁} E_ℙ((r⁰ + r′ζ̃)⁺) ≤ π¹(r⁰, r)`. -/
theorem theorem_1_upper {n : ℕ} (V Vhat : Set (Fin n → ℝ)) (r0 : ℝ) (r : Fin n → ℝ) :
TractableDRO.MeanSupport.worstCase (TractableDRO.MeanSupport.family1 V Vhat) r0 r ≤ pi1 V Vhat r0 r := by sorry
end TractableDRO.Unified
Source
Goh & Sim, Distributionally Robust Optimization and Its Tractable Approximations, Oper. Res. 58(4), 2010, p. 909, Theorem 1, (21) (upper-bound part)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.