Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 6.2 (proof), p. 106 — Lemma E.2 for F_M − f^{M̃} at γ/4: E_µ[f^M(π_M) − f^{M̃}(π_M)] ≤ 2Δ + 24θ log²(γ ∨ e)/γ + (γ/4)E[(f^M − f^{M̃})²]

Proved
StatComplexityDM.Disagreement.ps_fixed_shift_bound

by mikedeng1 · Oct 8, 2026 · Mathlib 0df444a (Lean v4.33.1)

decouplingdisagreement-coefficientp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-bookp2o-v1posterior-sampling

Let FM\mathcal{F}_{\mathcal{M}}FM​ be a class of mean-reward functions Π→[0,1]\Pi \to [0, 1]Π→[0,1], and let μ=∑iwiδMi\mu = \sum_i w_i \delta_{M_i}μ=∑i​wi​δMi​​ be a finitely supported prior over models with fMi∈FMf^{M_i} \in \mathcal{F}_{\mathcal{M}}fMi​∈FM​ and πMi∈arg⁡max⁡πfMi(π)\pi_{M_i} \in \arg\max_{\pi} f^{M_i}(\pi)πMi​​∈argmaxπ​fMi​(π). Let p=ρμp = \rho_\mup=ρμ​, ρμ(π):=μ({πM=π})\rho_\mu(\pi) := \mu(\{\pi_M = \pi\})ρμ​(π):=μ({πM​=π}), be the posterior sampling distribution. Fix a model M~\widetilde{M}M with fM~∈FMf^{\widetilde{M}} \in \mathcal{F}_{\mathcal{M}}fM∈FM​, and let γ≥e\gamma \ge eγ≥e. Then for every Δ>0\Delta > 0Δ>0,

EM∼μ[fM(πM)−fM~(πM)]≤2Δ+24 θ(FM−fM~,Δ,γ−1;ρμ)log⁡2(γ∨e)γ+γ4⋅Eπ∼p, M∼μ[(fM(π)−fM~(π))2],\mathbb{E}_{M\sim\mu}\bigl[f^M(\pi_M) - f^{\widetilde{M}}(\pi_M)\bigr] \le 2\Delta + 24\, \frac{\theta(\mathcal{F}_{\mathcal{M}} - f^{\widetilde{M}}, \Delta, \gamma^{-1}; \rho_\mu) \log^2(\gamma \vee e)}{\gamma} + \frac{\gamma}{4} \cdot \mathbb{E}_{\pi\sim p,\, M\sim\mu}\Bigl[\bigl(f^M(\pi) - f^{\widetilde{M}}(\pi)\bigr)^2\Bigr],EM∼μ​[fM(πM​)−fM(πM​)]≤2Δ+24γθ(FM​−fM,Δ,γ−1;ρμ​)log2(γ∨e)​+4γ​⋅Eπ∼p,M∼μ​[(fM(π)−fM(π))2],

where π∼p\pi \sim pπ∼p and M∼μM \sim \muM∼μ are independent and θ\thetaθ is the disagreement coefficient of Definition 6.3.

This is Lemma E.2 applied to the shifted class FM−fM~⊆(Π→[−1,1])\mathcal{F}_{\mathcal{M}} - f^{\widetilde{M}} \subseteq (\Pi \to [-1, 1])FM​−fM⊆(Π→[−1,1]) with parameter γ/4\gamma/4γ/4; averaging it over M~∼μ\widetilde{M} \sim \muM∼μ gives Theorem 6.2.

Formalization Note The infimum over Δ>0\Delta > 0Δ>0 is stated as a bound for every Δ>0\Delta > 0Δ>0. The prior and ρμ\rho_\muρμ​ are finite weighted families; models enter only through their mean rewards and decisions. The page derives this display inside the proof of Theorem 6.2, whose standing range is γ≥e\gamma \ge eγ≥e.

Preamble
import Mathlib
import Definitions.Def_StatComplexityDM_Disagreement_Coefficient
Formal statement
namespace StatComplexityDM.Disagreement

/-- Proof of Theorem 6.2, p. 106 (arXiv:2112.13487v3), Lemma E.2 applied with the class `F_M − f^{M̃}`
and parameter `γ/4`: for a class `F_M ⊆ (Π → [0, 1])`, a finitely supported prior
`µ = Σ_i w i · δ_{M_i}` with `fM i = f^{M_i} ∈ F_M` and `piM i = π_{M_i}`, posterior sampling
`p = ρ_µ = Σ_j w j · δ_{piM j}`, a fixed `M̃ ∈ M` with mean reward `g = f^{M̃} ∈ F_M`, `γ ≥ e` and
every `Δ > 0`,
`E_{M∼µ}[f^M(π_M) − f^{M̃}(π_M)] ≤ 2Δ + 24 θ(F_M − f^{M̃}, Δ, γ⁻¹; ρ_µ) log²(γ ∨ e)/γ
  + (γ/4) · E_{π∼p, M∼µ}[(f^M(π) − f^{M̃}(π))²]`. -/
theorem ps_fixed_shift_bound {ι Act : Type*} [Fintype ι] (FM : Set (Act → ℝ))
    (hF : ∀ f ∈ FM, ∀ π, 0 ≤ f π ∧ f π ≤ 1) (w : ι → ℝ) (hw : StatComplexityDM.LowerBound.IsDist w) (fM : ι → Act → ℝ)
    (hfM : ∀ i, fM i ∈ FM) (piM : ι → Act) (hpiM : ∀ i π, fM i π ≤ fM i (piM i))
    (g : Act → ℝ) (hg : g ∈ FM) (γ : ℝ) (hγ : Real.exp 1 ≤ γ) (Δ : ℝ) (hΔ : 0 < Δ) :
    ∑ i, w i * (fM i (piM i) - g (piM i))
      ≤ 2 * Δ + 24 * disagreementCoeff (shiftClass FM g) Δ γ⁻¹ w piM
          * Real.log (max γ (Real.exp 1)) ^ 2 / γ
        + γ / 4 * ∑ j, w j * ∑ i, w i * (fM i (piM j) - g (piM j)) ^ 2 := by sorry

end StatComplexityDM.Disagreement
Source
arXiv:2112.13487v3, App. E.2.2, proof of Theorem 6.2, p. 106 ("We now apply Lemma E.2 with the class F_M − f^M̃ and parameter γ/4")
Human review
  • Endorsed by Shuze Chen · Oct 9, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 9, 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