Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 1 — an extremizing bid-pair sequence with limiting area ≤c\le c≤c gives lim inf⁡W/O≥1−c/ζ−ϵ/ζ\liminf W/O\ge 1-c/\zeta-\epsilon/\zetaliminfW/O≥1−c/ζ−ϵ/ζ

Proved
LargeSupplyFPA.Asymptotic.theorem_1

by Shuze Chen · Nov 10, 2026 · Mathlib 0df444a (Lean v4.33.1)

bayes-nash-equilibriumfirst-price-auctionmulti-unit-auctionp2o-gran-per-chapterp2o-plan-paperp2o-v1price-of-anarchy

Fix H>0H>0H>0 and κ>0\kappa>0κ>0. There is a constant ζ>0\zeta>0ζ>0, depending only on HHH and κ\kappaκ, with the following property. Consider any sequence of kkk-unit first-price auctions A(k)A^{(k)}A(k) (with n(k)>kn(k)>kn(k)>k agents) satisfying Assumption 1 with these H,κH,\kappaH,κ, and any corresponding sequence E(k)E^{(k)}E(k) of ϵ(k)\epsilon^{(k)}ϵ(k)-approximate Bayes–Nash equilibria satisfying Assumption 3, with ϵ:=lim⁡kϵ(k)\epsilon:=\lim_k\epsilon^{(k)}ϵ:=limk​ϵ(k). Let c≥0c\ge0c≥0, and let {b‾(k),b‾(k)}\{\underline b^{(k)},\overline b^{(k)}\}{b​(k),b(k)} be an extremizing sequence of bid pairs along a subsequence (A(kl))l(A^{(k_l)})_l(A(kl​))l​ such that

lim⁡l→∞∫b‾(kl)b‾(kl)B(kl)(b) db≤c.\lim_{l\to\infty}\int_{\underline b^{(k_l)}}^{\overline b^{(k_l)}}B^{(k_l)}(b)\,db\le c.l→∞lim​∫b​(kl​)b(kl​)​B(kl​)(b)db≤c.

Then

lim inf⁡l→∞W(E(kl),A(kl))O(A(kl)) ≥ 1−cζ−ϵζ.\liminf_{l\to\infty}\frac{W(E^{(k_l)},A^{(k_l)})}{O(A^{(k_l)})}\ \ge\ 1-\frac{c}{\zeta}-\frac{\epsilon}{\zeta}.l→∞liminf​O(A(kl​))W(E(kl​),A(kl​))​ ≥ 1−ζc​−ζϵ​.

This is the asymptotic smoothness framework. The limiting area under the critical-bid distribution between an extremizing pair, plus the best-response slack, controls the limiting welfare loss.

Formalization Note ζ\zetaζ is chosen after H,κH,\kappaH,κ and before the sequence, n(⋅)n(\cdot)n(⋅), ϵ\epsilonϵ and ccc. The paper's proof gives O(A(k))≥kζO(A^{(k)})\ge k\zetaO(A(k))≥kζ with ζ=κH2/2\zeta=\kappa H^2/2ζ=κH2/2. The ratio lies in [0,1][0,1][0,1], so the real Filter.liminf is the genuine one.

Differences from the source:

  • encoding: a strategy is a map E_i : V_i → Δ(ℝ≥0) from values to bid distributions → a Markov kernel E i : Kernel ℝ ℝ with E i v ((−∞, 0)) = 0 for every v; reason: measurability in v is what makes the marginal bid distribution ∫ E_{i,v} dV_i defined; a law on ℝ with no mass below 0 is a law on ℝ≥0
  • encoding: a sequence of k-unit auctions (A^(k))_{k∈ℕ} with n(k) > k agents; subsequences (A^(k_l)) → n : ℕ → ℕ with k < n k, agents Fin (n k), value laws V k and kernels E k indexed by the unit count k (k = 0 included); a subsequence is a strictly increasing φ : ℕ → ℕ and the auction φ l has φ l units; reason: Lean indexing; the term k = 0 does not affect any limit
  • reading: Assumption 1(a): every value is at most H almost surely (values are non-negative in the model; 1(b) is stated on [0, H]) → V_i([0, H]) = 1 together with the density bound of 1(b), and 0 < H, 0 < κ; reason: the paper's values are non-negative (its bound O(A^(k)) ≥ kζ and W ≥ 0 use it, and p. 6 fixes H > 0, κ > 0); under Assumption 3 alone non-negativity is not implied, so it is read into Assumption 1
  • reading: E_{b∼E_{i,v}}[u_i(v, b)] ≥ sup_{b≥0} u_i(v, b) − ϵ for every v ∈ supp(V_i) → for every v ∈ supp(V_i): b ↦ u_i(v, b) is E_{i,v}-integrable and u_i(v, b) − ϵ ≤ ∫ u_i(v, ·) dE_{i,v} for every b ≥ 0; reason: the ∀ b form is equivalent to the sup form; integrability makes explicit that the paper's expectation is finite (it holds automatically when bids are bounded on the support, as under Assumptions 2 and 3)
  • encoding: Let ϵ := lim_{k→∞} ϵ^(k) → a hypothesis Tendsto εs atTop (𝓝 ε); reason: the limit is assumed to exist
  • encoding: a bid pair {b̲^(k), b̄^(k)} with b̲^(k) ≤ b̄^(k) for every auction (bids are non-negative) → lo, hi : ℕ → ℝ with 0 ≤ lo k ≤ hi k for every k (inside IsExtremizing); reason: bids live in ℝ≥0
  • reading: ζ > 0 is a constant that depends on the parameters of the environment → ∃ ζ > 0 quantified after H, κ and before the sequence, n(·), ϵ, c and the bid pairs; reason: p. 6 names H and κ as the parameters of the environment; a ζ depending on the sequence would make the bound vacuous
  • encoding: lim_{l→∞} ∫{b̲^(k_l)}^{b̄^(k_l)} B^(k_l)(b) db ≤ c (the limit exists) → ∃ L, Tendsto (∫{b̲^(k_l)}^{b̄^(k_l)} B^(k_l)(b) db) atTop (𝓝 L) ∧ L ≤ c; reason: the printed limit is assumed to exist
Preamble
import Mathlib
import Definitions.Def_LargeSupplyFPA_Asymptotic_Model
import Definitions.Def_LargeSupplyFPA_Asymptotic_Assumptions
import Definitions.Def_LargeSupplyFPA_Asymptotic_Extremizing

open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace LargeSupplyFPA.Asymptotic

theorem theorem_1 (H κ : ℝ) (hH : 0 < H) (hκ : 0 < κ) :
    ∃ ζ : ℝ, 0 < ζ ∧
      ∀ (n : ℕ → ℕ) (V : ∀ k, Fin (n k) → Measure ℝ) (E : ∀ k, Fin (n k) → Kernel ℝ ℝ)
        (εs : ℕ → ℝ) (ε : ℝ),
        IsAuctionSequence n V E →
        (∀ k, Assumption1 H κ (V k)) →
        (∀ k, IsApproxBNE k (V k) (E k) (εs k)) →
        Assumption3 n V E →
        Tendsto εs atTop (𝓝 ε) →
        ∀ (c : ℝ), 0 ≤ c →
        ∀ (lo hi : ℕ → ℝ) (φ : ℕ → ℕ), IsExtremizing n V E lo hi φ →
        (∃ L : ℝ, Tendsto (fun l => ∫ b in lo (φ l)..hi (φ l),
            kthBidCDF (φ l) (V (φ l)) (E (φ l)) b) atTop (𝓝 L) ∧ L ≤ c) →
        1 - c / ζ - ε / ζ ≤
          Filter.liminf (fun l => welfareRatio (φ l) (V (φ l)) (E (φ l))) atTop := by sorry

end LargeSupplyFPA.Asymptotic
Source
Hartline, Shah, vonAllmen, Robust Analysis of First-Price Auctions with Large Supply, STOC 2027 submission (manuscript of November 3, 2026, sha256 2ad8363b…), p. 11, Theorem 1 (proof pp. 12–14)
Human review
  • Endorsed by Community (Bot) · Nov 10, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Shuze Chen · Nov 10, 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