Theorem 5.3 with the explicit bound |h_β(i)| ≤ MK
ProvedArapostathisAC.RossRecurrence.theorem_5_3_explicitdiscounted-costp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1relative-value-functionreturn-time
Consider the countable-state controlled Markov process of §5 with bounded cost, for every and . Let and assume Ross's recurrence condition (5.7): there is with for all and . Then the differential discounted value function satisfies
This is the bound the proof of Theorem 5.3 produces ("The desired result follows from (5.9) and (5.10)"). The bound depends only on the cost bound and the recurrence constant .
Formalization Note. The paper's Theorem 5.3 asserts only that is bounded uniformly; this item states the explicit constant obtained in its proof, which is a stronger statement. Under bounded cost is finite, so is a genuine difference of real numbers.
Preamble
import Mathlib import Definitions.Def_ArapostathisAC_RossRecurrence_CMP import Definitions.Def_ArapostathisAC_RossRecurrence_ReturnTime open MeasureTheory ProbabilityTheory open scoped ENNReal
Formal statement
namespace ArapostathisAC.RossRecurrence
/-- Theorem 5.3 with the bound its proof gives (p. 302, "The desired result follows from (5.9) and
(5.10)"): under bounded costs `c ≤ M₀` and (5.7), `|h_β(i)| ≤ M₀ K` for all `β ∈ (0, 1)` and all
states `i`. -/
theorem theorem_5_3_explicit {A : Type*} [MetricSpace A] [MeasurableSpace A] [BorelSpace A]
(M : CMP A) (M₀ : ℝ) (hc : ∀ i, ∀ a ∈ M.U i, M.c i a ≤ M₀)
(K : ℝ) (hK : 0 < K)
(h57 : ∀ f : StationaryPolicy M, ∀ i, meanReturnTime M f i < ENNReal.ofReal K) :
∀ β ∈ Set.Ioo (0 : ℝ) 1, ∀ i, |hRel M β i| ≤ M₀ * K := by sorry
end ArapostathisAC.RossRecurrence
Source
Arapostathis, Borkar, Fernández-Gaucherand, Ghosh, Marcus, Discrete-time controlled Markov processes with average cost criterion: a survey, SIAM J. Control Optim. 31(2) (1993), p. 302, proof of Theorem 5.3, conclusion from (5.9) and (5.10)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.