Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 4.6 (corrected), stability step with tameness: box expansion for truncated ranks of tame modules yields an ε-multi-bijection of the truncated diagrams

Open
PersistClust.Count.theorem_4_5_lemma_4_6_corrected_stability_tame

by fabianroll · Oct 9, 2026 · Mathlib 0df444a (Lean v4.33.1)

clusteringpersistencetopological-data-analysis

This is the tameness-corrected form of the stability step of Lemma 4.6 of the paper (p. 21; proof in Appendix A), the algebraic-stability ingredient of Theorem 4.5.

Let XXX and YYY be tame 0-dimensional persistence modules, encoded by stage families stageX,stageY:R→Set(ι)\mathrm{stage}_X, \mathrm{stage}_Y : \mathbb{R} \to \mathrm{Set}(\iota)stageX​,stageY​:R→Set(ι) decreasing in the threshold, and component relations JX,JYJ_X, J_YJX​,JY​. Tameness means that the stages eventually vanish: ∃s0 ∀s≥s0, stage(s)=∅\exists s_0\, \forall s \ge s_0,\ \mathrm{stage}(s) = \varnothing∃s0​∀s≥s0​, stage(s)=∅. Write rX=rankFn⁡(stageX,JX)r_X = \operatorname{rankFn}(\mathrm{stage}_X, J_X)rX​=rankFn(stageX​,JX​) and rY=rankFn⁡(stageY,JY)r_Y = \operatorname{rankFn}(\mathrm{stage}_Y, J_Y)rY​=rankFn(stageY​,JY​) for the rank functions, and r~X=truncRank⁡(rX,α)\widetilde r_X = \operatorname{truncRank}(r_X, \alpha)rX​=truncRank(rX​,α), r~Y=truncRank⁡(rY,α)\widetilde r_Y = \operatorname{truncRank}(r_Y, \alpha)rY​=truncRank(rY​,α) for their α\alphaα-truncations (the rank functions restricted to the quadrant [α,∞)2[\alpha, \infty)^2[α,∞)2).

Assume the persistence diagrams of rXr_XrX​ and rYr_YrY​ have finite support, and that the truncated rank functions satisfy the box-expansion inequalities with no α\alphaα-guard: for all s,t∈Rs, t \in \mathbb{R}s,t∈R with t+2ε≤st + 2\varepsilon \le st+2ε≤s,

r~X(s,t)≤r~Y(s−ε, t+ε),r~Y(s,t)≤r~X(s−ε, t+ε).\widetilde r_X(s, t) \le \widetilde r_Y(s - \varepsilon,\, t + \varepsilon), \qquad \widetilde r_Y(s, t) \le \widetilde r_X(s - \varepsilon,\, t + \varepsilon).rX​(s,t)≤rY​(s−ε,t+ε),rY​(s,t)≤rX​(s−ε,t+ε).

Then there exists a multi-bijection δ\deltaδ between the copies of the two truncated diagrams mult⁡(r~X)\operatorname{mult}(\widetilde r_X)mult(rX​) and mult⁡(r~Y)\operatorname{mult}(\widetilde r_Y)mult(rY​) that moves every point by at most ε\varepsilonε in the L∞L^\inftyL∞ norm, in both coordinates and both directions.

The tameness hypothesis is necessary: without it, +∞+\infty+∞-born classes are invisible to the diagram (the off-diagonal multiplicity guard discards points with infinite birth coordinate) while still inflating the truncated rank functions, so the box inequalities can hold for rank functions whose truncated diagrams are fundamentally incomparable (one empty, the other not) — a machine-checked counterexample exists for the untamed sibling statement. With tameness (satisfied by the superlevel and Rips modules appearing in Theorem 4.5), the Extended-Stability argument of Appendix A applies.

Preamble
import Mathlib
import Definitions.Def_PersistClust_Count_Diagram
import Definitions.Def_PersistClust_Count_FiltrationLaw
import Definitions.Def_PersistClust_Count_TruncRank
Formal statement
namespace PersistClust.Count

theorem theorem_4_5_lemma_4_6_corrected_stability_tame
    {ιX : Type*} (stageX : ℝ → Set ιX) (JX : ℝ → ιX → ιX → Prop)
    {ιY : Type*} (stageY : ℝ → Set ιY) (JY : ℝ → ιY → ιY → Prop)
    (α ε : ℝ) (hε : 0 ≤ ε)
    (hLawX : FiltrationLaw stageX JX) (hLawY : FiltrationLaw stageY JY)
    (hTameX : ∃ s₀ : ℝ, ∀ s ≥ s₀, stageX s = ∅)
    (hTameY : ∃ s₀ : ℝ, ∀ s ≥ s₀, stageY s = ∅)
    (hbox : ∀ s t : ℝ, t + 2 * ε ≤ s →
      truncRank (rankFn stageX JX) α s t ≤ truncRank (rankFn stageY JY) α (s - ε) (t + ε) ∧
      truncRank (rankFn stageY JY) α s t ≤ truncRank (rankFn stageX JX) α (s - ε) (t + ε))
    (hDiagX : IsDiagramLike (mult (rankFn stageX JX)))
    (hDiagY : IsDiagramLike (mult (rankFn stageY JY)))
    (hFinX : {p : EReal × EReal | (mult (rankFn stageX JX)) p ≠ 0}.Finite)
    (hFinY : {p : EReal × EReal | (mult (rankFn stageY JY)) p ≠ 0}.Finite) :
    ∃ δ : Copies (mult (truncRank (rankFn stageX JX) α))
        ≃ Copies (mult (truncRank (rankFn stageY JY) α)),
      (∀ a, closeE (pt a).1 (pt (δ a)).1 ε ∧ closeE (pt a).2 (pt (δ a)).2 ε) ∧
      (∀ b, closeE (pt (δ.symm b)).1 (pt b).1 ε ∧ closeE (pt (δ.symm b)).2 (pt b).2 ε) := by sorry

end PersistClust.Count
Source
Chazal–Guibas–Oudot–Skraba, Persistence-Based Clustering in Riemannian Manifolds, INRIA RR-6968, 2009, pp. 20–21, Lemma 4.6 and Appendix A; algebraic stability: Chazal–Cohen-Steiner–Glisse–Oudot, https://hal.inria.fr/inria-00160683

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