Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Inequality (5), p. 8 — ES∼μ′[tr⁡(∑i∈Svivi⊤)−1]≤αtr⁡(∑ixivivi⊤)−1\mathbb E_{\mathcal S\sim\mu'}[\operatorname{tr}(\sum_{i\in\mathcal S}v_iv_i^\top)^{-1}]\le\alpha\operatorname{tr}(\sum_i x_iv_iv_i^\top)^{-1}ES∼μ′​[tr(∑i∈S​vi​vi⊤​)−1]≤αtr(∑i​xi​vi​vi⊤​)−1

Proved
PropVolSampling.Reduction.inequality_5

by mikedeng1 · Oct 20, 2026 · Mathlib 0df444a (Lean v4.33.1)

approximation-algorithmsoptimal-designp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1volume-sampling

Let 1≤d≤k≤n1\le d\le k\le n1≤d≤k≤n be integers and V=[v1,…,vn]V=[v_1,\dots,v_n]V=[v1​,…,vn​] a real d×nd\times nd×n matrix. Let x∈[0,1]nx\in[0,1]^nx∈[0,1]n with ∑ixi=k\sum_ix_i=k∑i​xi​=k be such that ∑ixivivi⊤\sum_ix_iv_iv_i^\top∑i​xi​vi​vi⊤​ is positive definite, and let μ\muμ be a probability measure on U≤k\mathcal U_{\le k}U≤k​ that is α\alphaα-approximate (d−1,d)(d-1,d)(d−1,d)-wise independent with respect to xxx, with positive normaliser Z=∑Sμ(S)det⁡(VSVS⊤)Z=\sum_S\mu(S)\det(V_SV_S^\top)Z=∑S​μ(S)det(VS​VS⊤​). If μ′\mu'μ′ is proportional volume sampling with measure μ\muμ, then

ES∼μ′[tr⁡(∑i∈Svivi⊤)−1]≤α tr⁡(∑i∈[n]xivivi⊤)−1.\mathbb E_{\mathcal S\sim\mu'}\Big[\operatorname{tr}\Big(\sum_{i\in\mathcal S}v_iv_i^\top\Big)^{-1}\Big]\le\alpha\,\operatorname{tr}\Big(\sum_{i\in[n]}x_iv_iv_i^\top\Big)^{-1}.ES∼μ′​[tr(i∈S∑​vi​vi⊤​)−1]≤αtr(i∈[n]∑​xi​vi​vi⊤​)−1.

Applied to an optimal solution xxx of the relaxation (1)–(3), the right-hand side is α CP≤α OPT\alpha\,\mathrm{CP}\le\alpha\,\mathrm{OPT}αCP≤αOPT; this is the core of Theorem 1.3.

Formalization Note Approximate independence is in the cross-multiplied form of Definition 1.2. Positivity of ZZZ is assumed so that μ′\mu'μ′ is a probability distribution, and positive definiteness of ∑ixivivi⊤\sum_ix_iv_iv_i^\top∑i​xi​vi​vi⊤​ so that the right-hand side is the page's finite value rather than Lean's junk tr⁡(0)\operatorname{tr}(0)tr(0). 1≤d1\le d1≤d is added because d−1d-1d−1 is natural-number subtraction.

Preamble
import Mathlib
import Definitions.Def_PropVolSampling_Reduction_Model
Formal statement
namespace PropVolSampling.Reduction
open Matrix Finset

theorem inequality_5 {d n : ℕ} (V : Matrix (Fin d) (Fin n) ℝ) (k : ℕ) (hd : 1 ≤ d) (hdk : d ≤ k)
    (hkn : k ≤ n) (x : Fin n → ℝ) (hx : IsRelaxFeasible k x) (hpd : (weightedGram V x).PosDef)
    (μ : Finset (Fin n) → ℝ) (hμ : IsSetDist k μ) (α : ℝ) (hα : IsApproxIndep d α x μ)
    (hZ : 0 < pvsNormalizer V μ) :
    pvsExpect V μ (fun S => aObj (gram V S)) ≤ α * aObj (weightedGram V x) := by sorry

end PropVolSampling.Reduction
Source
Nikolov, Singh, Tantipongpipat, Proportional Volume Sampling and Approximation Algorithms for A-Optimal Design, arXiv:1802.08318v5, p. 8, inequality (5) (proved on pp. 9–10 via (7)–(9))
Human review
  • Endorsed by Shuze Chen · Oct 20, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 20, 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