Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 5.1 — every ddd-dimensional geometric class is learnable with O(d2log⁡n+dk)O(d^2\log n+dk)O(d2logn+dk) queries by a deterministic learner that does not know kkk

Open
SortingCounterexamples.Optimal.Supplement.theorem_5_1

by mikedeng1 · Nov 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

geometric-rankingslearning-theoryp2o-batch-stoc27sp2o-gran-per-chapterp2o-plan-paperp2o-v1query-complexityrankingssorting

There is an absolute constant C>0C>0C>0 with the following property. Let n≥2n\ge2n≥2 and d≥0d\ge0d≥0, let v1,…,vn∈Rdv_1,\dots,v_n\in\mathbb R^dv1​,…,vn​∈Rd, and let HHH be a finite class of rankings of the nnn items such that every π∈H\pi\in Hπ∈H is induced by some direction w∈Rdw\in\mathbb R^dw∈Rd (that is, i≺πj  ⟺  ⟨vi,w⟩<⟨vj,w⟩i\prec_\pi j\iff\langle v_i,w\rangle<\langle v_j,w\ranglei≺π​j⟺⟨vi​,w⟩<⟨vj​,w⟩). Then there is a deterministic learner AAA, which may depend on nnn, ddd, HHH and vvv but not on kkk, whose queries are arbitrary linear orders of the items (not necessarily in HHH), such that for every k≥0k\ge0k≥0, every target σ∈H\sigma\in Hσ∈H and every adaptive oracle whose counterexamples to AAA include at most kkk untruthful ones for σ\sigmaσ, AAA obtains approval after

Q(A) ≤ C (d2log⁡n+dk)Q(A)\ \le\ C\,(d^2\log n+dk)Q(A) ≤ C(d2logn+dk)

queries, the approved query included.

This is the upper half of Theorem 1.2: in a low-dimensional geometric class the number of queries depends on the dimension ddd and only logarithmically on nnn.

Formalization Note The query count lives in {1,2,… }∪{∞}\{1,2,\dots\}\cup\{\infty\}{1,2,…}∪{∞}, with ∞\infty∞ when the oracle never approves, so the bound also asserts termination. For d=0d=0d=0 and n≥2n\ge2n≥2 no strict ranking is induced and the statement is vacuous, as in the source.

Differences from the source:

  • reading: O(d² log n + dk) for n items, no explicit domain → n ≥ 2, every d and every k ≥ 0; reason: at n = 1 the rate d² log 1 + dk vanishes for k = 0 while one query is always made; the base goal uses the same domain
  • encoding: can be learned using O(d² log n + dk) queries → ∃ C > 0 before n, d, v, H; a learner chosen after H and v and before k, σ and the oracle with Q ≤ ofReal (C · (d² log n + d k)); reason: this is the meaning of an O bound uniform in all parameters for a learner that does not know k; natural logarithm
  • encoding: at most k untruthful counterexamples → every transcript of the run of A against the oracle has at most k lies for σ; reason: this is exactly every adaptive response path with at most k lies against the deterministic learner
Preamble
import Mathlib
import Definitions.Def_SortingCounterexamples_Optimal_Protocol
import Definitions.Def_SortingCounterexamples_Optimal_Supplement_Geometric

open scoped ENNReal
Formal statement
namespace SortingCounterexamples.Optimal.Supplement

open SortingCounterexamples.Optimal

/-- Theorem 5.1: there is an absolute `C > 0` such that for every `n ≥ 2`, every `d`, every
`d`-dimensional geometric class `H` of rankings of `n` items with representation `v`, there is
one deterministic learner (its queries are arbitrary linear orders, not necessarily in `H`),
chosen before and independently of `k`, which for every `k`, every target `σ ∈ H` and every
oracle returning at most `k` untruthful counterexamples along the run obtains approval within
`C (d² log n + d k)` queries. -/
theorem theorem_5_1 :
    ∃ C : ℝ, 0 < C ∧ ∀ (n d : ℕ), 2 ≤ n → ∀ (v : Fin n → Fin d → ℝ) (H : Finset (Ranking n)),
      IsGeometricClass v H → ∃ A : DetLearner n, ∀ k : ℕ, ∀ σ ∈ H, ∀ O : Oracle n,
        RunAdmissible A O σ k →
          ((queries A O : ℕ∞) : ℝ≥0∞) ≤ ENNReal.ofReal (C * upperRate n d k) := by sorry

end SortingCounterexamples.Optimal.Supplement
Source
Noga Alon, Shay Moran, Shlomo Moran, STOC 2027 submission #549 (October 2026), printed p. 22 / PDF p. 23, Theorem 5.1 (proof to printed p. 25 / PDF p. 26); upper half of Theorem 1.2, printed p. 4 / PDF p. 5; model printed p. 1 / PDF p. 2

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