Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 1(a) — synchronous inexact proximal BR reaches an ϵ\epsilonϵ-NE2_22​ within ℓi(η)\ell_i(\eta)ℓi​(η) projected SG steps per player

Proved
LeiBR.Sync.theorem_1a

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

best-responseiteration-complexityp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1stochastic-approximationstochastic-nash-game

Consider the NNN-player stochastic Nash game under Assumptions 1 and 2, with μ>0\mu>0μ>0, a Nash equilibrium x∗x^*x∗, and Qi=2Mi2/μ2+2DXi2Q_i=2M_i^2/\mu^2+2D_{X_i}^2Qi​=2Mi2​/μ2+2DXi​2​. Run Algorithm 1 from a deterministic x0∈Xx_0\in Xx0​∈X with ∥xi,0−xi∗∥≤C\|x_{i,0}-x_i^*\|\le C∥xi,0​−xi∗​∥≤C, computing each inexact proximal BR by the inner loop (SAi,k_{i,k}i,k​) with

ji,k=⌈Qiη2(k+1)⌉j_{i,k}=\Big\lceil\frac{Q_i}{\eta^{2(k+1)}}\Big\rceilji,k​=⌈η2(k+1)Qi​​⌉

projected stochastic gradient steps at major iteration kkk, for some η∈(0,1)\eta\in(0,1)η∈(0,1) (so that (8) holds with αi,k=ηk+1\alpha_{i,k}=\eta^{k+1}αi,k​=ηk+1). The sampled gradients satisfy the hypotheses of Lemma 3 at every (k,i,t)(k,i,t)(k,i,t) with 1≤t≤ji,k1\le t\le j_{i,k}1≤t≤ji,k​, and the sampled gradient map and the samples are measurable.

Let a=∥Γ∥a=\|\Gamma\|a=∥Γ∥, c=max⁡{a,η}c=\max\{a,\eta\}c=max{a,η}, q∈(c,1)q\in(c,1)q∈(c,1), D=1/ln⁡((q/c)e)D=1/\ln((q/c)^e)D=1/ln((q/c)e), 0<ϵ<N(C+D)0<\epsilon<\sqrt N(C+D)0<ϵ<N​(C+D), and run

K=⌈ln⁡(N(C+D)/ϵ)ln⁡(1/q)⌉K=\Big\lceil\frac{\ln\big(\sqrt N(C+D)/\epsilon\big)}{\ln(1/q)}\Big\rceilK=⌈ln(1/q)ln(N​(C+D)/ϵ)​⌉

major iterations. Then:

  1. xKx_KxK​ is an ϵ\epsilonϵ-NE2_22​ in the sense of (26): E∥(∥x1,K−x1∗∥,…,∥xN,K−xN∗∥)∥≤ϵ\mathbb E\big\|(\|x_{1,K}-x_1^*\|,\dots,\|x_{N,K}-x_N^*\|)\big\|\le\epsilonE​(∥x1,K​−x1∗​∥,…,∥xN,K​−xN∗​∥)​≤ϵ;
  2. for every player iii, the number of projected stochastic gradient steps used, ∑k=0K−1ji,k\sum_{k=0}^{K-1}j_{i,k}∑k=0K−1​ji,k​, is at most
ℓi(η)=Qiη4ln⁡(1/η2)(N(C+D)ϵ)ln⁡(1/η2)ln⁡(1/q)+⌈ln⁡(N(C+D)/ϵ)ln⁡(1/q)⌉.\ell_i(\eta)=\frac{Q_i}{\eta^4\ln(1/\eta^2)}\left(\frac{\sqrt N(C+D)}{\epsilon}\right)^{\frac{\ln(1/\eta^2)}{\ln(1/q)}}+\left\lceil\frac{\ln\big(\sqrt N(C+D)/\epsilon\big)}{\ln(1/q)}\right\rceil .ℓi​(η)=η4ln(1/η2)Qi​​(ϵN​(C+D)​)ln(1/q)ln(1/η2)​+⌈ln(1/q)ln(N​(C+D)/ϵ)​⌉.

This is the paper's overall iteration complexity of the synchronous scheme: an ϵ\epsilonϵ-Nash equilibrium is reached with a number of sampled gradients per player that is explicit in ϵ\epsilonϵ, NNN, η\etaη and the game's constants.

Formalization Note The theorem states exactly what the proof establishes, (29) and the bound after (31), for the step counts ji,kj_{i,k}ji,k​ and the iteration count KKK used there. Statement repairs: the initial point is deterministic (Algorithm 1), so E∥xi,0−xi∗∥≤C\mathbb E\|x_{i,0}-x_i^*\|\le CE∥xi,0​−xi∗​∥≤C reads ∥xi,0−xi∗∥≤C\|x_{i,0}-x^*_i\|\le C∥xi,0​−xi∗​∥≤C; ϵ<N(C+D)\epsilon<\sqrt N(C+D)ϵ<N​(C+D) is assumed, since otherwise the ceiling in (27) can be non-positive while no step is taken; the second-moment hypothesis of Lemma 3 is the bound Mi2M_i^2Mi2​; Assumption 1(b) is joint C2C^2C2. D=1/(eln⁡(q/c))D=1/(e\ln(q/c))D=1/(eln(q/c)).

Preamble
import Mathlib
import Definitions.Def_SpectralProjGrad_Shared_IsProjOnto
import Definitions.Def_LeiBR_Sync_NashGame
import Definitions.Def_LeiBR_Sync_StochGame
import Definitions.Def_LeiBR_Sync_Algorithm1

open MeasureTheory
Formal statement
namespace LeiBR.Sync

/-- Theorem 1(a), p. 11 (overall iteration complexity of the synchronous scheme). Run Algorithm 1
with the inner loop (SA_{i,k}), `j_{i,k} = ⌈Q_i/η^{2(k+1)}⌉` projected stochastic-gradient steps
for player `i` at major iteration `k` (so `α_{i,k} = η^{k+1}`, `η ∈ (0, 1)`), from a
deterministic `x_0 ∈ X` with `‖x_{i,0} - x*_i‖ ≤ C`, for
`K = ⌈ln(√N(C+D)/ϵ)/ln(1/q)⌉` major iterations, where `a = ‖Γ‖`, `c = max{a, η}`,
`q ∈ (c, 1)`, `D = 1/ln((q/c)^e)` and `0 < ϵ < √N(C+D)`. Then `x_K` is an `ϵ`-NE₂,
`E‖(‖x_{i,K} - x*_i‖)_i‖ ≤ ϵ`, and for every player the total number of steps
`∑_{k<K} j_{i,k}` is at most
`ℓ_i(η) = Q_i/(η⁴ ln(1/η²)) · (√N(C+D)/ϵ)^{ln(1/η²)/ln(1/q)} + ⌈ln(√N(C+D)/ϵ)/ln(1/q)⌉`. -/
theorem theorem_1a
    {N : ℕ} {n : Fin N → ℕ} {Ω : Type*} [mΩ : MeasurableSpace Ω] (P : Measure Ω)
    [IsProbabilityMeasure P] {d : ℕ} (ξ : Ω → EuclideanSpace ℝ (Fin d))
    (X : ∀ i : Fin N, Set (Strat n i)) (ψ : Fin N → Profile n → EuclideanSpace ℝ (Fin d) → ℝ)
    (gψ : ∀ i : Fin N, Profile n → EuclideanSpace ℝ (Fin d) → Strat n i) (M : Fin N → ℝ)
    (hA1 : Assumption1 P ξ X ψ gψ M) (μ : ℝ) (hμ : 0 < μ)
    (hA2 : Assumption2 X (payoff P ξ ψ) μ)
    (xhat : Profile n → Profile n) (hxhat : IsProxBR X (payoff P ξ ψ) μ xhat)
    (proj : ∀ i : Fin N, Strat n i → Strat n i)
    (hproj : ∀ i : Fin N, SpectralProjGrad.Shared.IsProjOnto (X i) (proj i))
    (hgψ : ∀ i : Fin N,
      Measurable (fun p : Profile n × EuclideanSpace ℝ (Fin d) => gψ i p.1 p.2))
    (j : Fin N → ℕ → ℕ) (ξs : ℕ → Fin N → ℕ → Ω → EuclideanSpace ℝ (Fin d))
    (hξs : ∀ k (i : Fin N) t, Measurable (ξs k i t))
    (x0 : Profile n) (hx0 : x0 ∈ stratSet X)
    (xs : Profile n) (hxs : IsNashEq X (payoff P ξ ψ) xs)
    (η : ℝ) (hη0 : 0 < η) (hη1 : η < 1)
    (hj : ∀ (i : Fin N) (k : ℕ), j i k = ⌈Qconst X M μ i / η ^ (2 * (k + 1))⌉₊)
    (hint : ∀ k (i : Fin N) (t : ℕ), 1 ≤ t → t ≤ j i k →
      Integrable (fun ω => ‖gψ i (Function.update (syncSA proj gψ μ j ξs x0 k ω) i
        (saZ proj gψ μ j ξs x0 k i t ω)) (ξs k i t ω)‖ ^ 2) P)
    (hunb : ∀ k (i : Fin N) (t : ℕ), 1 ≤ t → t ≤ j i k →
      P[fun ω => gψ i (Function.update (syncSA proj gψ μ j ξs x0 k ω) i
          (saZ proj gψ μ j ξs x0 k i t ω)) (ξs k i t ω) | innerField j ξs k i t]
        =ᵐ[P] fun ω => partialGrad (payoff P ξ ψ) i
          (Function.update (syncSA proj gψ μ j ξs x0 k ω) i (saZ proj gψ μ j ξs x0 k i t ω)))
    (hvar : ∀ k (i : Fin N) (t : ℕ), 1 ≤ t → t ≤ j i k →
      P[fun ω => ‖gψ i (Function.update (syncSA proj gψ μ j ξs x0 k ω) i
          (saZ proj gψ μ j ξs x0 k i t ω)) (ξs k i t ω)‖ ^ 2 | innerField j ξs k i t]
        ≤ᵐ[P] fun _ => M i ^ 2)
    (C : ℝ) (hC : ∀ i : Fin N, ‖x0 i - xs i‖ ≤ C)
    (q : ℝ) (hcq : max (specNorm (Gamma X (payoff P ξ ψ) μ)) η < q) (hq1 : q < 1)
    (D : ℝ) (hD : D = 1 / (Real.exp 1 *
      Real.log (q / max (specNorm (Gamma X (payoff P ξ ψ) μ)) η)))
    (ϵ : ℝ) (hϵ0 : 0 < ϵ) (hϵ1 : ϵ < Real.sqrt N * (C + D))
    (K : ℕ) (hK : K = ⌈Real.log (Real.sqrt N * (C + D) / ϵ) / Real.log (1 / q)⌉₊) :
    ∫ ω, blockDist (syncSA proj gψ μ j ξs x0 K ω) xs ∂P ≤ ϵ ∧
      ∀ i : Fin N, ∑ k ∈ Finset.range K, (j i k : ℝ) ≤
        Qconst X M μ i / (η ^ 4 * Real.log (1 / η ^ 2)) *
            (Real.sqrt N * (C + D) / ϵ) ^ (Real.log (1 / η ^ 2) / Real.log (1 / q)) +
          (⌈Real.log (Real.sqrt N * (C + D) / ϵ) / Real.log (1 / q)⌉₊ : ℝ) := by sorry

end LeiBR.Sync
Source
Lei, Shanbhag, Pang & Sen, On Synchronous, Asynchronous, and Randomized Best-Response Schemes for Stochastic Nash Games, arXiv:1704.04578v2, p. 11, Theorem 1(a), (26)–(27); proof p. 12, (29)–(31)
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 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