Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 7.3.3 — T and T_f are contractions of modulus βα_b on the bounding-function norm

Proved
MDPFinance.Contracting.lemma_7_3_3

by Shuze Chen · Sep 27, 2026 · Mathlib 0df444a (Lean v4.33.1)

banach-spacecontraction-mappingmarkov-decision-process

Once the model has a genuine bounding function bbb (a two-sided reward bound, Definition 7.3.1), both operators TfT_fTf​ and TTT are Lipschitz on (IBb,∥⋅∥b)(IB_b, \|\cdot\|_b)(IBb​,∥⋅∥b​) with constant βαb\beta\alpha_bβαb​ — a contraction exactly when βαb<1\beta\alpha_b < 1βαb​<1. Banach's fixed point theorem then gives, for a fixed decision rule fff, that JfJ_fJf​ (the value of the stationary policy f∞f^\inftyf∞) is the unique fixed point of TfT_fTf​ in IBbIB_bIBb​, obtained as the norm-limit of iterating TfT_fTf​ from any starting point. This is the technical engine behind every result of §7.3: contraction is what turns the abstract existence theory of §7.1-7.2 into concrete, computable, uniquely-determined solutions.

Preamble
import Mathlib
import Definitions.Def_MDPFinance_Contracting_Model
import Definitions.Def_MDPFinance_Contracting_Value
import Definitions.Def_MDPFinance_Contracting_Bounding

open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace MDPFinance.Contracting

/-- Lemma 7.3.3 (Bäuerle–Rieder, p. 206, PDF 217). Suppose the Markov Decision Model has a
bounding function `b` and let `f \in F`. a) For `v,w \in IB_b` it holds: `\|T_fv-T_fw\|_b \le
\beta\alpha_b\|v-w\|_b`, `\|Tv-Tw\|_b \le \beta\alpha_b\|v-w\|_b`. b) Let `\beta\alpha_b < 1`.
Then `J_f = \lim_n T_f^ng` for all `g \in IB_b` (norm convergence), and `J_f` is the unique fixed
point of `T_f` in `IB_b`; `J_f` renders `x \mapsto (J_{\infty}^{f^\infty}(x)).toReal`, real-valued
under the contracting hypothesis. -/
theorem lemma_7_3_3 {E A : Type*} [MeasurableSpace E] [MeasurableSpace A] (M : MarkovDecisionModel E A)
    (b : E → ℝ) (cr αb : ℝ) (hb : IsBoundingFunction M b cr αb) (f : E → A)
    (hf : IsDecisionRuleOf M f) :
    (∀ v ∈ IBb b, ∀ w ∈ IBb b,
        normb b (fun x => Tf' M f v x - Tf' M f w x) ≤ M.β * αb * normb b (fun x => v x - w x)) ∧
      (∀ v ∈ IBb b, ∀ w ∈ IBb b,
        normb b (fun x => T' M v x - T' M w x) ≤ M.β * αb * normb b (fun x => v x - w x)) ∧
      (M.β * αb < 1 →
        (∀ g ∈ IBb b, Tendsto (fun n => normb b fun x =>
            (Tf' M f)^[n] g x - (Jinfpi M M.r (fun _ => f) x).toReal) atTop (𝓝 0)) ∧
          ∀ v ∈ IBb b, Tf' M f v = v → v = fun x => (Jinfpi M M.r (fun _ => f) x).toReal) := by sorry

end MDPFinance.Contracting
Source
Bäuerle and Rieder, Markov Decision Processes with Applications to Finance, Universitext, Springer 2011, DOI 10.1007/978-3-642-18324-9, p. 206, Lemma 7.3.3
Human review
  • Endorsed by Community (Bot) · Oct 1, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Shuze Chen · Oct 1, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me