pith. sign in

arxiv: 2211.05981 · v4 · pith:2HIQATJMnew · submitted 2022-11-11 · 🧮 math.AT · math.RT

Multiparameter persistence modules in the large scale

Martin Frankland , Donald Stanley This is my paper

Pith reviewed 2026-05-24 10:53 UTC · model grok-4.3

classification 🧮 math.AT math.RT
keywords multiparameter persistence modulesindecomposable diagramslarge-scale equivalencewild classificationposet representationsvector space diagramsalgebraic topology
0
0 comments X

The pith

Two-dimensional multiparameter persistence modules have their indecomposables classified up to finitely supported diagrams, but the problem is wild in higher dimensions.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper studies persistence modules with multiple discrete parameters, which are diagrams of vector spaces indexed by the poset of m-tuples of natural numbers. To focus on large-scale behavior, diagrams are considered equivalent if they agree outside a negligible region. In the two-dimensional case, the indecomposable such diagrams are classified completely once finitely supported diagrams are quotiented out. In higher dimensions the authors give a partial classification up to suitably finite diagrams and prove that the full problem of classifying indecomposables is wild.

Core claim

The authors establish that for two parameters the indecomposable diagrams of vector spaces over the two-dimensional grid are classifiable modulo finitely supported diagrams under the large-scale equivalence, while for three or more parameters the classification problem remains wild even after restricting attention to diagrams that are finite outside a negligible region.

What carries the argument

The equivalence relation that identifies diagrams agreeing outside a negligible region, applied to diagrams of vector spaces over the poset N^m to capture large-scale multiparameter persistence.

If this is right

  • In two dimensions a complete list of indecomposable types exists up to finite support.
  • The classification problem in dimensions three and higher is representation-theoretically wild.
  • Partial lists of indecomposables can still be obtained in higher dimensions by restricting to suitably finite diagrams.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach may extend to other posets beyond the grid of natural numbers.
  • Wildness in higher dimensions implies that computational methods for multiparameter persistence may need to focus on special cases or invariants rather than full decomposition.
  • Large-scale equivalence could be useful in applications where small-scale noise is irrelevant.

Load-bearing premise

The chosen notion of equivalence using negligible regions accurately reflects the large-scale behavior intended for study.

What would settle it

Finding an indecomposable diagram in dimension two that lies outside the claimed classification list, or exhibiting a finite classification in dimension three, would disprove the claims.

Figures

Figures reproduced from arXiv: 2211.05981 by Donald Stanley, Martin Frankland.

Figure 1
Figure 1. Figure 1: Some indecomposable objects of L(K2). Lemma 7.2. For any ~a ∈ N m, there is an isomorphism in L(Km) [~a,∞) ∼= LKm(t ~aR), where t ~a = t a1 1 · · ·t am m . Lemma 7.3. Let M be a finitely generated Rσi -module, for some 1 ≤ i ≤ m. Let x ∈ M be a (non-zero) element of minimal degree in the product partial order, that is: |x|i ≤ |y|i for all y ∈ M, and let hxi ⊆ M denote the Rσi -submodule generated by x. The… view at source ↗
Figure 2
Figure 2. Figure 2: The “scanning” process in the proof of Lemma 7.5. Lemma 7.7. Let M be an object in F [PITH_FULL_IMAGE:figures/full_fig_p016_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The “battleship game” in the proof of Lemma 7.9. Lemma 7.10. In the case m = 2, let M be a torsion-free module in F. Then any epimorphism q : M ։ [~a,∞) constructed as in Lemma 7.9 is split. Proof. Write yi := t ~a−ai~ei t ai i 1 ∈ [ai ,∞)i for the canonical generator, shifted to bidegree ~a for convenience. Those agree in the terminal corner: ϕ1(y1) = ϕ2(y2) = y = t ~a1 ∈ k[t ± 1 , t± 2 ] = R[2] = [~a,∞)[… view at source ↗
Figure 4
Figure 4. Figure 4: The modules M and N in Example 10.2. The modules M and N have the same rank invariant: rkM(~a,~b) =    0 if ~a = (0, 0) 1 if ~a = (a, 0) or ~a = (0, a) for some a > 0 2 if ~a ≥ (1, 1) [PITH_FULL_IMAGE:figures/full_fig_p025_4.png] view at source ↗
read the original abstract

A persistence module with $m$ discrete parameters is a diagram of vector spaces indexed by the poset $\mathbb{N}^m$. If we are only interested in the large scale behavior of such a diagram, then we can consider two diagrams equivalent if they agree outside of a ``negligeable'' region. In the $2$-dimensional case, we classify the indecomposable diagrams up to finitely supported diagrams. In higher dimension, we partially classify the indecomposable diagrams up to suitably finite diagrams, and show that the full classification problem is wild.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript introduces an equivalence relation on N^m-indexed persistence modules (diagrams of vector spaces) under which two modules are identified if they agree outside a 'negligible' region. It claims a classification of indecomposable diagrams up to finitely supported diagrams in the two-parameter case, a partial classification up to suitably finite diagrams in higher dimensions, and a proof that the unrestricted classification problem is wild.

Significance. A rigorous classification of indecomposables under a well-motivated large-scale equivalence would be a useful contribution to multiparameter persistence theory, where the standard classification problem is known to be wild. The two-dimensional result, if complete and if the equivalence is shown to preserve relevant invariants, could provide a concrete starting point for constructing large-scale invariants; the wildness statement in higher dimensions would clarify the boundary of tractability.

major comments (1)
  1. [Abstract] Abstract: the central classification statements are stated with respect to equivalence 'outside a negligible region,' yet the abstract provides no derivation or comparison showing that this relation coincides with the kernels of natural maps that forget finite-scale data (e.g., via interleaving distance or stability). Without such a link, it is unclear whether the classified indecomposables are precisely those that survive at infinity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive feedback. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central classification statements are stated with respect to equivalence 'outside a negligible region,' yet the abstract provides no derivation or comparison showing that this relation coincides with the kernels of natural maps that forget finite-scale data (e.g., via interleaving distance or stability). Without such a link, it is unclear whether the classified indecomposables are precisely those that survive at infinity.

    Authors: We agree that the abstract would benefit from an explicit link to standard stability notions. The manuscript defines the equivalence directly via agreement outside a finitely supported (negligible) region to isolate large-scale behavior, but does not derive it as the kernel of an interleaving or stability map in the abstract itself. In revision we will expand the abstract with one sentence noting that the relation is coarser than, but compatible with, finite interleavings and therefore classifies modules that are indistinguishable at infinity. This addresses the concern without altering the technical results. revision: yes

Circularity Check

0 steps flagged

No significant circularity; classification is self-contained

full rationale

The paper defines an equivalence on N^m-indexed diagrams (agreement outside a negligible region) explicitly to study large-scale behavior, then classifies indecomposables under that equivalence using standard methods from poset representation theory. No derivation step reduces to a fitted parameter renamed as prediction, no self-citation is load-bearing for the central claim, and the equivalence is not smuggled in via prior work by the same authors. The 2D classification and higher-dimensional wildness result follow from the chosen equivalence without tautological reduction to inputs. This is a normal non-circular finding for a direct classification paper.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on the standard definition of persistence modules as functors from the poset N^m to vector spaces and on the usual notions of indecomposability and equivalence of diagrams; no free parameters or new entities are introduced in the abstract.

axioms (2)
  • standard math Persistence modules are diagrams of vector spaces indexed by the poset N^m
    This is the standard definition used throughout multiparameter persistence literature.
  • standard math Indecomposable objects are those that cannot be written as direct sums of nonzero objects
    Standard definition in representation theory of posets.

pith-pipeline@v0.9.0 · 5606 in / 1250 out tokens · 27973 ms · 2026-05-24T10:53:43.968833+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

13 extracted references · 13 canonical work pages

  1. [1]

    [AS05] D. M. Arnold and D. Simson, Endo-wild representation type and generic representation s of finite posets , Pacific J. Math. 219 (2005), no. 1, 1–26, DOI 10.2140/pjm.2005.219.1. MR217421 8 [ABE+22] H. Asashiba, M. Buchet, E. G. Escolar, K. Nakashima, and M . Yoshiwaki, On interval decom- posability of 2D persistence modules , Comput. Geom. 105 (2022), ...

    work page doi:10.2140/pjm.2005.219.1 2005

  2. [2]

    Bauer, M

    MR0242802 [BBOS20] U. Bauer, M. B. Botnan, S. Oppermann, and J. Steen, Cotorsion torsion triples and the representation the- ory of filtered hierarchical clustering, Adv. Math. 369 (2020), 107171, 51, DOI 10.1016/j.aim.2020.107171. MR4091895 [BBH22] B. Blanchette, T. Br¨ ustle, and E. J. Hanson, Homological approximations in persistence theory (2022), Prep...

    work page doi:10.1016/j.aim.2020.107171 2020

  3. [3]

    [BCB20] M. B. Botnan and W. Crawley-Boevey, Decomposition of persistence modules , Proc. Amer. Math. Soc. 148 (2020), no. 11, 4581–4596, DOI 10.1090/proc/14790. MR4143 378 [BLO20a] M. B. Botnan, V. Lebovici, and S. Oudot, On rectangle-decomposable 2-parameter persistence modul es, 36th International Symposium on Computational Geometry, L IPIcs. Leibniz In...

    work page doi:10.1090/proc/14790 2020

  4. [4]

    [BOO22] M

    MR4117735 [BLO20b] , Local characterizations for decomposability of 2-paramet er persistence modules (2020), Preprint, available at arXiv:2008.02345. [BOO22] M. B. Botnan, S. Oppermann, and S. Oudot, Signed barcodes for multi-parameter persistence via rank decompositions, 38th International Symposium on Computational Geometry, LIPIcs. Leibniz Int. Proc. I...

    work page doi:10.4230/lipics.socg.2022.19 2020

  5. [5]

    B¨ uhler,Exact categories , Expo

    MR33 63157 [B¨ uh10] T. B¨ uhler,Exact categories , Expo. Math. 28 (2010), no. 1, 1–69, DOI 10.1016/j.exmath.2009.04.004. MR2606234 [CZ09] G. Carlsson and A. Zomorodian, The theory of multidimensional persistence , Discrete Comput. Geom. 42 (2009), no. 1, 71–93, DOI 10.1007/s00454-009-9176-0. MR25 06738 [CCS21] W. Chach´ olski, R. Corbet, and A.-L. Sattel...

    work page doi:10.1016/j.exmath.2009.04.004 2010

  6. [6]

    [Cox95] D

    (2023), available at arXiv:2303.08270. [Cox95] D. A. Cox, The homogeneous coordinate ring of a toric variety , J. Algebraic Geom. 4 (1995), no. 1, 17–50. MR1299003 [CLS11] D. A. Cox, J. B. Little, and H. K. Schenck, Toric varieties, Graduate Studies in Mathematics, vol. 124, American Mathematical Society, Providence, RI,

    work page arXiv 2023

  7. [7]

    MR3727119 [Dic66] S. E. Dickson, A torsion theory for Abelian categories , Trans. Amer. Math. Soc. 121 (1966), 223–235, DOI 10.2307/1994341. MR191935 [EH18] K. Erdmann and T. Holm, Algebras and representation theory , Springer Undergraduate Mathematics Series, Springer, Cham,

    work page doi:10.2307/1994341 1966

  8. [8]

    Gabriel, Des cat´ egories ab´ eliennes, Bull

    MR3837168 [Gab62] P. Gabriel, Des cat´ egories ab´ eliennes, Bull. Soc. Math. France 90 (1962), 323–448 (French). MR232821 [GC21] O. G¨ afvert and W. Chach´ olski,Stable Invariants for Multiparameter Persistence (2021), Preprint, avail- able at arXiv:1703.03632. [GM03] S. I. Gelfand and Y. I. Manin, Methods of homological algebra , 2nd ed., Springer Monog...

    work page arXiv 1962

  9. [9]

    MR1950475 (2003m: 18001) [HOST19] H. A. Harrington, N. Otter, H. Schenck, and U. Tillm ann, Stratifying multiparameter persistent homology, SIAM J. Appl. Algebra Geom. 3 (2019), no. 3, 439–471, DOI 10.1137/18M1224350. MR4000203 [Har77] R. Hartshorne, Algebraic geometry, Springer-Verlag, New York-Heidelberg,

    work page doi:10.1137/18m1224350 2019

  10. [10]

    Kim and S

    MR0463157 [KM21] W. Kim and S. Moore, The Generalized Persistence Diagram Encodes the Bigraded B etti Numbers (2021), Preprint, available at arXiv:2111.02551. [Les15] M. Lesnick, The theory of the interleaving distance on multidimensiona l persistence modules , Found. Comput. Math. 15 (2015), no. 3, 613–650, DOI 10.1007/s10208-015-9255-y. MR 3348168 [Naz7...

    work page doi:10.1007/s10208-015-9255-y 2021

  11. [11]

    Oudot and L

    MR3408277 [OS22] S. Oudot and L. Scoccola, On the stability of multigraded Betti numbers and Hilbert fu nctions (2022), Preprint, available at arXiv:2112.11901. [Per04] M. Perling, Graded rings and equivariant sheaves on toric varieties , Math. Nachr. 263/264 (2004), 181– 197, DOI 10.1002/mana.200310130. MR2029751 [Per13] , Resolutions and cohomologies of...

    work page doi:10.1002/mana.200310130 2022

  12. [12]

    Shah, Krull-Remak-Schmidt decompositions in Hom-finite additiv e categories, Expo

    M R0340375 [Sha23] A. Shah, Krull-Remak-Schmidt decompositions in Hom-finite additiv e categories, Expo. Math. 41 (2023), no. 1, 220–237, DOI 10.1016/j.exmath.2022.12.003. MR4557 280 [Sim05] D. Simson, On Corner type Endo-Wild algebras , J. Pure Appl. Algebra 202 (2005), no. 1-3, 118–132, DOI 10.1016/j.jpaa.2005.01.012. MR2163404 [SS07a] D. Simson and A. S...

    work page doi:10.1016/j.exmath.2022.12.003 2023

  13. [13]

    Vipond, Multiparameter persistence landscapes , J

    [Vip20] O. Vipond, Multiparameter persistence landscapes , J. Mach. Learn. Res. 21 (2020), Paper No. 61,

    work page 2020