Corollary 7.5.4 — increasing plus a standard policy gives (SEN)
ProvedSennottDP.BOR.sen_of_monotone_valueaverage-costp2o-batch-b23ap2o-gran-per-chapterp2o-plan-bookp2o-v1queueing
Let and assume that is increasing in for every . If there exists a standard (randomized stationary) policy, then the (SEN) assumptions hold for the distinguished state . Moreover every limit function is nonnegative and increasing in :
This is the route to (SEN3) through a structural property of the discount value function.
Preamble
import Mathlib import Definitions.Def_SennottDP_BOR_Assumptions open scoped ENNReal NNReal open Filter Topology
Formal statement
namespace SennottDP.BOR
/-- Sennott (1999), Corollary 7.5.4, p. 144. Let `S = {0, 1, 2, …}` and assume `V_α` is
increasing in `i` for every `α ∈ (0,1)`. If there exists a `0` standard (randomized stationary)
policy, then the (SEN) assumptions hold (for the distinguished state `0`), and every limit function
is nonnegative and increasing in `i`. -/
theorem sen_of_monotone_value {Act : Type} (M : SennottDP.Discounted.MDC ℕ Act)
(hmono : ∀ α : ℝ, 0 < α → α < 1 → Monotone (valueFn M α))
(hd : ∃ d : RandStationaryPolicy M, IsZStandard d.toPolicy 0) :
SEN M 0 ∧ ∀ h : ℕ → ℝ, IsLimitFunction M 0 h → (∀ i, 0 ≤ h i) ∧ Monotone h := by sorry
end SennottDP.BOR
Source
Sennott, Stochastic Dynamic Programming and the Control of Queueing Systems (Wiley, 1999), p. 144, Corollary 7.5.4
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.