Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 8.4.3 — the unichain optimality equation has a solution, with a unique gain

Proved
MarkovDecisionProcesses.average_optimality_equation_solvable

by naimengye · Sep 23, 2026 · Mathlib 0df444a (Lean v4.33.1)

dynamic-programmingmarkov-decision-processesoperations-researchprobability

Consider a stationary MDP with finite state and action sets that is unichain: the transition matrix of every deterministic stationary policy consists of a single recurrent class plus a possibly empty set of transient states. Then

  • (a) there exist a scalar ggg and a function h:S→Rh:S\to\mathbb Rh:S→R solving the average reward optimality equation B(g,h)=0B(g,h)=0B(g,h)=0, that is, for every state sss,
0=max⁡a∈As{r(s,a)−g+∑j∈Sp(j∣s,a) h(j)−h(s)};0=\max_{a\in A_s}\Bigl\{r(s,a)-g+\sum_{j\in S}p(j\mid s,a)\,h(j)-h(s)\Bigr\};0=a∈As​max​{r(s,a)−g+j∈S∑​p(j∣s,a)h(j)−h(s)};
  • (b) the scalar is unique: if (g′,h′)(g',h')(g′,h′) is any other solution then g=g′g=g'g=g′.

This is Theorem 8.4.3. The source proves (a) by letting the discount factor tend to 111 along a subsequence on which the same stationary policy is discount optimal (Theorem 6.2.10 of mission II), expanding the discounted value as (1−λ)−1ge+h+o(1)(1-\lambda)^{-1}g e+h+o(1)(1−λ)−1ge+h+o(1), and passing to the limit in the discounted optimality equation; the solution obtained is the gain and bias of that policy. Part (b) is Theorem 8.4.1(c). The source notes that hhh is not unique, since (g,h+ke)(g,h+ke)(g,h+ke) is again a solution.

Formalization Note Bounded rewards (Assumption 8.0.2) are automatic for a finite model and are not a separate hypothesis. Unichain is the source's Section 8.3.1 classification, stated through accessibility and recurrence on each PdP_dPd​ as in the definition file. The state space is assumed nonempty, as every state space of the source is: on an empty SSS every pair (g,h)(g,h)(g,h) solves B(g,h)=0B(g,h)=0B(g,h)=0 vacuously, and part (b) would be false.

Preamble
import Definitions.Def_MarkovDecisionProcesses_AverageReward
Formal statement
namespace MarkovDecisionProcesses
theorem average_optimality_equation_solvable {S A : Type*} [Fintype S] [Nonempty S]
    [Fintype A] (M : StationaryMDP S A) (hM : IsUnichain M) :
    (∃ (g : ℝ) (h : S → ℝ), ∀ s, optimalityResidual M g h s = 0) ∧
      ∀ (g : ℝ) (h : S → ℝ) (g' : ℝ) (h' : S → ℝ),
        (∀ s, optimalityResidual M g h s = 0) →
          (∀ s, optimalityResidual M g' h' s = 0) → g = g' := by sorry
end MarkovDecisionProcesses
Source
Martin L. Puterman, Markov Decision Processes: Discrete Stochastic Dynamic Programming, Wiley 1994, https://doi.org/10.1002/9780470316887 — §8.4.2, printed p. 358 (PDF p. 374), Theorem 8.4.3: "Suppose S and A_s are finite Assumption 8.0.2 holds and the model is unichain. a. Then there exists a g ∈ R¹ and an h ∈ V for which 0 = max_{d∈D}{r_d − ge + (P_d − I)h}. b. If (g′, h′) is any other solution of the average reward optimality equation, then g = g′."
Human review
  • Endorsed by Shuze Chen · Sep 25, 2026

    Confirmed by the moderator at approval.

  • Endorsed by naimengye · Sep 25, 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