Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

§8.3 — N₂ ≥ (1/ε) ln(1/β) makes the FAST bound (8.5) at most β

Proved
ScenarioApproach.Fast.fast_bound_le_of_N2

by mikedeng1 · Oct 2, 2026 · Mathlib 0df444a (Lean v4.33.1)

fast-algorithmp2o-batch-b23ap2o-gran-per-chapterp2o-plan-bookp2o-v1sample-complexityscenario-approach

Let ε,β∈(0,1)\varepsilon,\beta\in(0,1)ε,β∈(0,1) and let d,N1,N2d,N_1,N_2d,N1​,N2​ be natural numbers. If

N2 ≥ 1εln⁡1β,N_2\ \ge\ \frac1\varepsilon\ln\frac1\beta,N2​ ≥ ε1​lnβ1​,

then the right-hand side of the FAST bound (8.5) is at most β\betaβ:

(1−ε)N2∑i=0d−1(N1i)εi(1−ε)N1−i ≤ β.(1-\varepsilon)^{N_2}\sum_{i=0}^{d-1}\binom{N_1}{i}\varepsilon^i(1-\varepsilon)^{N_1-i}\ \le\ \beta.(1−ε)N2​i=0∑d−1​(iN1​​)εi(1−ε)N1​−i ≤ β.

Combined with Theorem 8.5 this gives the sample-size rule of FAST: with N1=KdN_1=KdN1​=Kd first-stage scenarios and N2≥1εln⁡1βN_2\ge\frac1\varepsilon\ln\frac1\betaN2​≥ε1​lnβ1​ detuning scenarios, PN1+N2{V(νF∗,ℓF∗)>ε}≤β\mathbb P^{N_1+N_2}\{V(\nu^*_F,\ell^*_F)>\varepsilon\}\le\betaPN1​+N2​{V(νF∗​,ℓF∗​)>ε}≤β, so the total number of scenarios grows additively, Kd+1εln⁡1βKd+\frac1\varepsilon\ln\frac1\betaKd+ε1​lnβ1​, rather than multiplicatively in ddd and 1/ε1/\varepsilon1/ε.

Formalization Note The statement holds for every ddd and N1N_1N1​; the book's choice N1=KdN_1=KdN1​=Kd plays no role in the inequality. The exponent N1−iN_1-iN1​−i is natural-number subtraction, which only matters for i>N1i>N_1i>N1​, where (N1i)=0\binom{N_1}{i}=0(iN1​​)=0.

Preamble
import Mathlib
Formal statement
namespace ScenarioApproach.Fast

/-- The `N₂` rule of FAST (p. 95): for `ε, β ∈ (0, 1)`, if `N₂ ≥ (1/ε) ln(1/β)` then the
right-hand side of (8.5) is at most `β`, whatever `N₁` and `d`:
`(1-ε)^{N₂} ∑_{i=0}^{d-1} (N₁ choose i) ε^i (1-ε)^{N₁-i} ≤ β`. -/
theorem fast_bound_le_of_N2 (d N₁ N₂ : ℕ) (ε β : ℝ)
    (hε0 : 0 < ε) (hε1 : ε < 1) (hβ0 : 0 < β) (hβ1 : β < 1)
    (hN₂ : (1 / ε) * Real.log (1 / β) ≤ N₂) :
    (1 - ε) ^ N₂ *
        ∑ i ∈ Finset.range d, (N₁.choose i : ℝ) * ε ^ i * (1 - ε) ^ (N₁ - i) ≤ β := by sorry

end ScenarioApproach.Fast
Source
Campi & Garatti, Introduction to the Scenario Approach, SIAM/MOS 2018, DOI 10.1137/1.9781611975444, p. 95, §8.3 (Pros: Reduced sample size requirements)
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 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