Proof of Proposition 3.1, p. 437 β under (A2), an algebraic subgradient is positive: l(X) β₯ 0 for X β π³β
ProvedConvexRiskFn.Cont.algSubgradient_nonnegalgebraic-subgradientmonotonicityp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1positive-functionalrisk-function
Let be a real vector space with a partial order compatible with addition, and let satisfy (A2) (monotonicity). Let with , and let be an algebraic subgradient of at , i.e. for all . Then is positive:
This is the step of the proof of Proposition 3.1 where monotonicity enters.
Formalization Note The paper takes with proper; the statement only needs that is finite ( and ), which is the situation of the paper. No topology is needed.
Preamble
import Mathlib import Definitions.Def_ConvexRiskFn_Cont_Setting open Filter Topology
Formal statement
namespace ConvexRiskFn.Cont
theorem algSubgradient_nonneg {E : Type*} [AddCommGroup E] [Module β E] [PartialOrder E]
[IsOrderedAddMonoid E] (Ο : E β EReal) (h2 : A2 Ο) (Xbar : E)
(hdom : Xbar β ConvexRiskFn.Dual.dom Ο) (hbot : β₯ < Ο Xbar) (l : E ββ[β] β)
(hl : IsAlgSubgradient Ο Xbar l) :
β X : E, 0 β€ X β 0 β€ l X := by sorry
end ConvexRiskFn.Cont
Source
RuszczyΕski, Shapiro, Optimization of convex risk functions, Math. Oper. Res. 31 (2006), p. 437, proof of Proposition 3.1, second and third sentences
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.