Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 5.1: a SFABP satisfies the conservation laws (5.5)-(5.6), with equality for policies giving priority to states outside S

Proved
AllocationIndices.sfabp_conservation_laws

by naimengye · Sep 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

achievable-regionconservation-lawsmarkov-bandits

Lemma 5.1. There exist positive AiSA_i^SAiS​ (i∈S⊆Ei \in S \subseteq Ei∈S⊆E), having the interpretation (5.3), such that for any scheduling policy π\piπ, ∑i∈SAiSxiπ≥b(S)\sum_{i \in S} A_i^S x_i^\pi \ge b(S)∑i∈S​AiS​xiπ​≥b(S) for all S⊂ES \subset ES⊂E (5.5), ∑i∈EAiExiπ=b(E)\sum_{i \in E} A_i^E x_i^\pi = b(E)∑i∈E​AiE​xiπ​=b(E) (5.6), and such that equality holds in (5.5) if π\piπ is a policy that gives priority to bandits whose states are not in SSS over any bandits whose states are in SSS.

Formally: for nnn identical bandit processes on E={1,…,N}E = \{1, \dots, N\}E={1,…,N} with kernel PPP, discount factor a∈(0,1)a \in (0, 1)a∈(0,1), initial state-vector kkk, every policy π\piπ of the Bandit Algorithms model and every S⊆ES \subseteq ES⊆E: AiS>0A_i^S > 0AiS​>0 for i∈Si \in Si∈S; ∑i∈SAiSxiπ≥b(S)\sum_{i \in S} A_i^S x_i^\pi \ge b(S)∑i∈S​AiS​xiπ​≥b(S); ∑i∈Exiπ=1/(1−a)\sum_{i \in E} x_i^\pi = 1/(1-a)∑i∈E​xiπ​=1/(1−a) (which is (5.6), as AiE=1A_i^E = 1AiE​=1 and b(E)=1/(1−a)b(E) = 1/(1-a)b(E)=1/(1−a)); and if π\piπ gives priority to bandits whose states are not in SSS, then ∑i∈SAiSxiπ=b(S)\sum_{i \in S} A_i^S x_i^\pi = b(S)∑i∈S​AiS​xiπ​=b(S). Here b(S)=(1−a)−1∏j:kj∉SE[aTkjS]b(S) = (1-a)^{-1} \prod_{j : k_j \notin S} \mathbb{E}[a^{T^S_{k_j}}]b(S)=(1−a)−1∏j:kj​∈/S​E[aTkj​S​], the minimal cost identified on p. 120 (the printed formula has a sum in place of the product).

Preamble
import Definitions.Def_AllocationIndices_Achievable

open MeasureTheory ProbabilityTheory BanditAlgorithm Finset
Formal statement
namespace AllocationIndices

/-- **Lemma 5.1** (p. 120). For a SFABP of `n` identical bandits on the finite state space
`E = Fin N` with discount factor `a ∈ (0, 1)`, started from the state-vector `k`, the
coefficients `Aᵢ^S` of (5.3) are positive for `i ∈ S`, and for every scheduling policy `π`,
`∑_{i∈S} Aᵢ^S xᵢᵖ ≥ b(S)` for every `S ⊆ E` (5.5), `∑_{i∈E} Aᵢ^E xᵢᵖ = b(E) = 1/(1−a)` (5.6),
and equality holds in (5.5) whenever `π` gives priority to bandits whose states are not in `S`
over any bandits whose states are in `S`. -/
theorem sfabp_conservation_laws {N n : ℕ} (P : Kernel (Fin N) (Fin N)) [IsMarkovKernel P]
    {a : ℝ} (ha0 : 0 < a) (ha1 : a < 1) (k : Fin n → Fin N) (π : MarkovBanditPolicy n (Fin N))
    (S : Finset (Fin N)) :
    (∀ i ∈ S, 0 < conservationCoeff P a S i) ∧
    conservationBase P a k S ≤ ∑ i ∈ S, conservationCoeff P a S i * performance P π a k i ∧
    ∑ i, performance P π a k i = (1 - a)⁻¹ ∧
    (GivesPriority S π →
      ∑ i ∈ S, conservationCoeff P a S i * performance P π a k i = conservationBase P a k S) := by sorry

end AllocationIndices
Source
Gittins, Glazebrook and Weber, Multi-armed Bandit Allocation Indices, 2nd ed., Wiley 2011, doi:10.1002/9780470980033, §5.3 pp. 120-121, Lemma 5.1 with the argument preceding it ((5.3), the cost problem and its value)
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