Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§4, p. 10 — if all policies are proper, the partition S_1, S_2, … by distance from the goal is exhaustive

Proved
StableFA.DerivedMDP.partition_exhaustive

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1partitionproper-policystochastic-shortest-path

Let MMM be a finite nondiscounted Markov decision process with cost-free absorbing goal state 111 in which every policy is proper. Define the layers S1={1}S_1=\{1\}S1​={1}, Uk=⋃j<kSjU_k=\bigcup_{j<k}S_jUk​=⋃j<k​Sj​ and, for k≥2k\ge 2k≥2,

Sk={x  ∣  x∉Uk ∧ min⁡amax⁡y∈UkP(δ(x,a)=y)>0}.S_k=\Bigl\{x\;\Bigm|\;x\notin U_k\ \wedge\ \min_a\max_{y\in U_k}P(\delta(x,a)=y)>0\Bigr\}.Sk​={x​x∈/Uk​ ∧ amin​y∈Uk​max​P(δ(x,a)=y)>0}.

Then every state lies in some layer: for every non-goal state xxx there is an index kkk with x∈Skx\in S_kx∈Sk​.

The report attributes this to Bertsekas and Tsitsiklis [BT89]. It makes the layer index k(x)k(x)k(x), the "distance from the goal", well defined, and with it the notion of a self-weighted averager.

Formalization Note "All policies proper" quantifies over stationary policies (functions from states to admissible actions, p. 3). The goal state lies in S1S_1S1​ by definition, so only non-goal states are quantified.

Preamble
import Mathlib
import Definitions.Def_BertsekasSSPModel
import Definitions.Def_StableFA_DerivedMDP_Setting
Formal statement
namespace StableFA.DerivedMDP

theorem partition_exhaustive {n : ℕ} {C : Type} [Fintype C]
    (M : BertsekasSSPModel n C)
    (hproper : ∀ μ : Fin n → C, (∀ i, μ i ∈ M.U i) → IsProper M μ) :
    ∀ i : Fin n, ∃ k : ℕ, some i ∈ layer M k := by sorry

end StableFA.DerivedMDP
Source
Gordon, Stable Function Approximation in Dynamic Programming, Tech. Rep. CMU-CS-95-103, Carnegie Mellon University (1995), p. 10, §4, "Bertsekas and Tsitsiklis show that this partitioning is exhaustive"
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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