Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 3 — the functional-equation solution is the optimal return

Proved
DenardoDP.Contraction.theorem3

by mikedeng1 · Oct 5, 2026 · Mathlib 0df444a (Lean v4.33.1)

contraction-mappingdynamic-programmingp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1

Suppose Denardo's contraction and monotonicity assumptions both hold. For every policy δ\deltaδ, let vδv_\deltavδ​ be its return function, the unique fixed point of HδH_\deltaHδ​. The maximization operator AAA then has exactly one fixed point v∗∈Vv^*\in Vv∗∈V, and this solution is the optimal return function fff, defined pointwise over all policies:

Av∗=v∗,v∗(x)=f(x)=sup⁡δ∈Δvδ(x)(x∈Ω).Av^*=v^*,\qquad v^*(x)=f(x)=\sup_{\delta\in\Delta}v_\delta(x)\quad(x\in\Omega).Av∗=v∗,v∗(x)=f(x)=δ∈Δsup​vδ​(x)(x∈Ω).

The theorem identifies the solution of the functional equation with the best policy return at every state. Its monotonicity assumption matters: the paper gives a contraction that violates monotonicity for which the two functions can differ.

Formalization Note The policy operators and AAA are supplied as maps V→VV\to VV→V constrained by equations (1) and (3), recording the paper's bounded-range assumptions. The family vδv_\deltavδ​ is supplied through equation (2), which has a unique solution for each policy under contraction. The formal statement includes existence and uniqueness of the fixed point and uses a real least-upper-bound predicate for the supremum of policy returns. IsMaxOperator enforces nonempty decision sets wherever a state exists.

Preamble
import Mathlib
import Definitions.Def_DenardoDP_Contraction_Model
Formal statement
namespace DenardoDP.Contraction

/-- Theorem 3, p. 168: the unique functional-equation solution is the optimal return. -/
theorem theorem3 {Ω : Type*} {D : Ω → Type*}
    (h : (x : Ω) → D x → BFun Ω → ℝ)
    (H : ((x : Ω) → D x) → BFun Ω → BFun Ω)
    (A : BFun Ω → BFun Ω) (c : ℝ)
    (v : ((x : Ω) → D x) → BFun Ω)
    (hH : IsPolicyOperator h H) (hA : IsMaxOperator h A)
    (hc : ContractionAssumption h c) (hmono : MonotonicityAssumption H)
    (hv : ∀ δ, H δ (v δ) = v δ) :
    (∃! w : BFun Ω, A w = w) ∧
      ∀ w : BFun Ω, A w = w → IsOptimalReturn v w := by sorry

end DenardoDP.Contraction
Source
Denardo, Contraction Mappings in the Theory Underlying Dynamic Programming, SIAM Review 9(2) (1967), p. 168, Theorem 3, with definitions on pp. 166–167; https://doi.org/10.1137/1009030
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 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