Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Bipath Persistence as Zigzag Persistence

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper proves that every bipath persistence module is encoded by an infinite periodic zigzag module, making the arc code a complete invariant and giving bipath persistence the same algebraic stability as zigzag persistence.

desk verdict Clean covering-map reduction of bipath to zigzag persistence, with a real isometry theorem, but the key lemma is unproved—send it out and ask for the details. read the letter →

arxiv 2501.00322 v1 pith:QTKR3Z6E submitted 2024-12-31 math.AT math.RT

Ángel Javier Alonso (1) , Enhao Liu (2) ((1) Institute of Geometry , Graz University of Technology , Graz , Austria , (2) Department of Mathematics , Kyoto University , Kyoto
show 1 more author
Japan)
This is my paper · ORCID
classification math.ATmath.RT MSC 55N3116G20
keywords bipathpersistencezigzagarccodecoveringmapalgebraicstabilitybottleneckdistanceinterleavingfibered
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper shows that bipath persistence—the third and last setting in which persistence modules always decompose into interval modules—can be reduced to zigzag persistence. The reduction is a covering map from an infinite zigzag poset to the bipath poset: restricting a bipath module along it produces an infinite periodic zigzag module whose barcode, modulo a $\mathbb{Z}$-shift, is exactly the bipath module's arc code. Because the barcode of that zigzag module is a complete isomorphism invariant, the arc code is complete as well. The paper then defines interleaving and bottleneck distances on bipath modules through the zigzag, proving they coincide and thereby transferring algebraic stability from zigzag to bipath persistence. It also shows that only a finite slice of the zigzag, of size $O(K)$ for maximal chain length $K$, is needed to compute the arc code.

What carries the argument

The carrying mechanism is the covering map $\zeta : ZZ \to B$, which wraps the infinite zigzag poset $ZZ$ around the bipath poset $B$ periodically, and the restriction functor $R : \mathrm{Vec}^B \to \mathrm{Vec}^{ZZ}$ it induces. The map is modeled on the universal covering of the circle and sends each of the two chains of the bipath to alternating segments of the zigzag. Under $R$, every interval module on the bipath becomes a direct sum of zigzag interval modules indexed by $\mathbb{Z}$, so the barcode of $R(M)$ is periodic and the arc code of $M$ is exactly the orbit space of that barcode under the $\mathbb{Z}$-shift.

What would settle it

Exhibit two non-isomorphic bipath modules $M$ and $N$ whose restrictions $R(M)$ and $R(N)$ have identical zigzag barcodes; Theorem 2.3 says the barcode of the restriction is a complete invariant, so such a pair cannot exist. Equivalently, finding any pair of bipath modules with $d_I(M,N) \neq d_B(M,N)$ would refute Theorem 3.5.

Watch

Extended reading notes

Core claim

The central claim is Theorem 2.3: for any bipath module $M$, the barcode $B(R(M))$ of the infinite zigzag module obtained by restricting along the covering map is a complete invariant up to isomorphism, and the arc code $B(M)$ is in bijection with the $\mathbb{Z}$-orbits of $B(R(M))$. Each interval of the bipath—full, left, right, top, or bottom—lifts to a $\mathbb{Z}$-periodic family of zigzag intervals, so the infinite zigzag barcode is periodic and its orbit space recovers the arc code. From this, Theorem 3.5 concludes that the interleaving distance $d_I(M,N)$ equals the bottleneck distance $d_B(M,N)$ for all bipath modules, which is algebraic stability: modules that are $\epsilon$-interleaved have arc codes that are $\epsilon$-close in bottleneck distance.

Load-bearing premise

The load-bearing premise is that every pointwise finite-dimensional persistence module over a bipath poset decomposes into interval modules; if that classification ever failed, the arc code and the reduction to a finite zigzag slice would not be defined for arbitrary bipath modules.

Editorial extensions

If this is right

  • To decompose a bipath module, one needs only decompose the finite zigzag slice of size $O(K)$, so any zigzag decomposition algorithm—including fast algorithms for simplex-wise filtrations—can be applied to bipath persistence directly.
  • Bipath modules inherit the isometry theorem: $d_I(M,N) = d_B(M,N)$, so algebraic stability of zigzag persistence automatically becomes algebraic stability of bipath persistence.
  • The arc code is a complete isomorphism invariant: two bipath modules are isomorphic if and only if their arc codes coincide.
  • The fibered arc code of a 2-D persistence module is strictly more discriminating than the fibered barcode: it separates two non-isomorphic modules with equal fibered barcode, though it is not itself complete.
  • Epsilon-matchings between arc codes can be taken equivariantly with respect to the $\mathbb{Z}$-action and pair intervals of the same bipath type, giving the bottleneck distance a combinatorial description directly on arc codes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • One extension the authors leave implicit is a general recipe: any finite poset equipped with a periodic zigzag covering would inherit zigzag stability and zigzag algorithms, with bipath posets as the worked-out instance.
  • Because the arc code is a $\mathbb{Z}$-quotient of a periodic barcode, one could try to prove stability of the fibered arc code under interleavings of $\mathbb{R}^2$ modules; the paper does not state such a result.
  • The finite-slice theorem makes bipath decomposition cost $O(K)$ times the cost of decomposing one zigzag of length about $K$; comparing this with the matrix-algorithm route on random bipath modules would be a direct practical test.
  • The periodic-zigzag viewpoint connects bipath persistence to persistence modules over finite subsets of the circle, where decompositions into intervals and Jordan cells appear; whether the arc code extends to that setting is left open.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces a poset map ζ from the infinite zigzag poset ZZ to a bipath poset B, and studies the induced restriction functor R from bipath modules to zigzag modules. It claims (Theorem 2.3) that the barcode of R(M) is a complete invariant up to isomorphism and that the arc code of M is recovered as the quotient of this barcode by a free Z-action. It then defines interleaving and bottleneck distances for bipath modules via R, obtaining an isometry theorem (Theorem 3.5) from the known block isometry theorem for zigzag modules. It gives a finite-slice version for computation (Theorem 4.1), and introduces a fibered arc code for R2 persistence modules, with examples comparing it to the fibered barcode and the interval rank invariant.

Significance. If the main theorems hold, the paper's covering construction is a valuable conceptual and computational bridge: bipath persistence becomes a periodic zigzag persistence, which lets known zigzag algorithms and stability results apply directly. The proposed finite-slice reduction to a single finite zigzag is appealing, and the explicit comparison of the fibered arc code with the fibered barcode in Section 5 is useful. The paper is honest about its dependence on the Aoki-Escolar-Tada interval-decomposability theorem and about limitations of the new invariant (Example 5.3). The main reservation is that the key restriction computation Lemma 2.2 is asserted without proof, and the finite-slice uniqueness assertion in Theorem 4.1 is only sketched.

major comments (3)
  1. [Section 2.2, Lemma 2.2] This lemma is the load-bearing step for Theorem 2.3, yet it is stated without proof. Please supply a complete proof that: (a) ζ is a poset map and the preimage of each interval is exactly the claimed disjoint union of ZZ-intervals; (b) the restriction of the interval module kI to each summand has the stated interval support with identity internal maps; and (c) the resulting periodic intervals from different bipath interval types or parameters are pairwise non-isomorphic modulo the Z-action, so that the orbit map in Theorem 2.3 is well defined and injective. Boundary cases (j=0 for left intervals, j=n for right intervals) need explicit treatment. The sentence 'straightforward to verify' in Section 2.1 for the classification of bipath intervals should also be expanded or referenced.
  2. [Section 4, Theorem 4.1] The proof reduces to the sentence 'It is not difficult to see...' and the final assertion that each listed zigzag interval is an indecomposable summand of R'(kI) and not of any other R'(kJ). Since the finite slice truncates infinite periodic summands, the identification of which summand survives in ZZ' and the proof of uniqueness require a case-by-case check over all interval types and parameters. Please provide the complete argument; as written, the computational claim is not independently verifiable.
  3. [Section 3.4, Lemma 3.6] The orbit-replacement argument in the proof does not obviously produce a bijection: if an interval J outside the orbit of I is matched by τ to some translate of τ(I), then after the replacement two source intervals are matched to the same target interval. Since Lemma 3.6 is used to justify the equivalent bottleneck-distance definition on arc codes and the claim that matchings pair intervals of the same type, either give a correct proof (for instance, by matching orbit representatives and using Z-equivariance of the interleaving distance) or present the arc-code matching as a separate definition whose equivalence is proved directly.
minor comments (5)
  1. [Section 2.1 / Example 5.2] The global assumption in Section 2.1 is that n,m are greater than 1, but Example 5.2 says n=3, m=1; the displayed poset in that example corresponds to n=3, m=2. Please correct the parameters and ensure the nondegeneracy assumption is consistent.
  2. [Section 3.3, Lemma 3.4] The second displayed equality should read E(k[a,b)_ZZ)=k[a,b)_BL, not k[a,b)_ZZ.
  3. [Section 4, Theorem 4.1] In the left-interval bullet with j≠0, write µ_M(k[i,j]_◁) instead of µ_M([i,j]_◁) for consistency.
  4. [Section 2.2] The statement that R preserves direct sums 'since our persistence modules are pointwise finite-dimensional' is misleading: precomposition with ζ preserves direct sums for all persistence modules. Consider rephrasing.
  5. [Section 5, Example 5.2] The matrices defining M_λ are hard to parse in the present rendering; please typeset them with clear row and column labels.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the main claims use external theorems and an explicit covering construction; the stability transfer is definitional but transparent.

full rationale

The derivation chain is not circular. The interval decomposability of bipath modules is imported from the external theorem of Aoki, Escolar and Tada [3, Theorem 1.3], and the barcode existence for infinite zigzag modules from Botnan [12]. The central new content is the covering map ζ and the restriction functor R; Lemma 2.2 computes R(kI) by direct inspection of the definition of ζ, and Theorem 2.3 then follows from this lemma together with preservation of direct sums. Although the proof of Lemma 2.2 is omitted, that is a correctness/completeness gap rather than a circularity, since the lemma is a concrete computation, not an assumption of the conclusion. The bipath interleaving and bottleneck distances in Section 3.4 are explicitly defined as the corresponding zigzag distances of R(M) and R(N); Theorem 3.5 is therefore a transparent definitional transfer of the external zigzag block-isometry theorem (Theorem 3.2), not a derivation that presupposes the bipath statement. The self-citations to Asashiba–Liu [8] and Asashiba–Gauthier–Liu [7] appear only as related work and examples, and are not load-bearing for the main theorems. No fitted parameter is renamed as a prediction, no uniqueness theorem from the authors is invoked to force a choice, and no ansatz is smuggled in via citation. Overall, the paper is self-contained modulo clearly cited external results and contains no circular step.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

The paper introduces one new construction (the covering map) and otherwise relies on standard theorems in representation theory and persistence. No empirical parameters are fitted.

assumptions (5)
  • domain assumption Every pointwise finite-dimensional persistence module over a bipath poset decomposes into interval modules (Aoki-Escolar-Tada, [3, Theorem 1.3]).
    Used to define the arc code and to prove the covering theorem and the finite-slice computation; not proven in this paper.
  • domain assumption Infinite zigzag persistence modules admit a unique interval decomposition (Botnan, [12]).
    Assures the barcode of R(M) is well-defined and complete.
  • domain assumption Block isometry theorem for zigzag modules (Bjerkevik, [10, Theorem 4.18]; Botnan-Lesnick, [11]).
    The bipath isometry theorem (Theorem 3.5) is a direct consequence of this external stability result.
  • standard math Krull-Schmidt theorem for pointwise finite-dimensional persistence modules over finite posets.
    Justifies that the interval decomposition of M is unique, making the arc code a well-defined multiset.
  • domain assumption The classification of intervals of a bipath poset into full, left, right, top, and bottom types is exhaustive.
    Stated as 'straightforward to verify' in Section 2.1; underpins Lemma 2.2 and Theorem 4.1.
invented entities (1)
  • Covering map ζ: ZZ → B independent evidence
    purpose: Relates bipath modules to infinite zigzag modules via the restriction functor R; this is the main construction of the paper.
    Explicitly defined in Definition 2.1 and used throughout; its properties are the subject of Lemma 2.2. It is a fully specified mathematical object, not an unexplained postulate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bipath Persistence as Zigzag Persistence." pith.science (2026). https://pith.science/paper/QTKR3Z6E

@misc{pith2026250100322,
  author       = {Pith},
  title        = {Pith review of: Bipath Persistence as Zigzag Persistence},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QTKR3Z6E}},
  note         = {Machine review of arXiv:2501.00322}
}
read the original abstract

Persistence modules that decompose into interval modules are important in topological data analysis because we can interpret such intervals as the lifetime of topological features in the data. We can classify the settings in which persistence modules always decompose into intervals, by a recent result of Aoki, Escolar and Tada: these are standard single-parameter persistence, zigzag persistence, and bipath persistence. No other setting offers such guarantees. We show that a bipath persistence module can be decomposed via a closely related infinite zigzag persistence module, understood as a covering. This allows us to translate techniques of zigzag persistence, like recent advancements in its efficient computation by Dey and Hou, to bipath persistence. In addition, and again by the relation with the infinite zigzag, we can define an interleaving and bottleneck distance on bipath persistence. In turn, the algebraic stability of zigzag persistence implies the algebraic stability of bipath persistence.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. On preservation of relative resolutions for poset representations

    math.RT 2025-06 conditional novelty 6.0 of 10

    For aligned interior systems of a poset, induction preserves interval covers and interval resolutions of persistence modules, and interval resolution dimensions are preserved.

Reference graph

Works this paper leans on

25 extracted references · 17 canonical work pages · cited by 1 Pith paper

  1. [1]

    Decomposition of Zero-Dimensional Persistence Modules via Rooted Subsets

    Ángel Javier Alonso and Michael Kerber. Decomposition of Zero-Dimensional Persistence Modules via Rooted Subsets. Discrete & Computational Geometry, November 2024. doi: 10.1007/s00454-024-00700-7

    work page doi:10.1007/s00454-024-00700-7 2024

  2. [2]

    Probabilistic Analysis of Mul- tiparameter Persistence Decompositions into Intervals

    Ángel Javier Alonso, Michael Kerber, and Primoz Skraba. Probabilistic Analysis of Mul- tiparameter Persistence Decompositions into Intervals. LIPIcs, Volume 293, SoCG 2024, 293:6:1–6:19, 2024. doi:10.4230/LIPICS.SOCG.2024.6

    doi:10.4230/lipics.socg.2024.6 2024

  3. [3]

    Summand-injectivity of interval covers and monotonicity of interval resolution global dimensions

    Toshitaka Aoki, Emerson G. Escolar, and Shunsuke Tada. Summand-Injectivity of Interval Covers and Monotonicity of Interval Resolution Global Dimensions, November 2023.arXiv: 2308.14979

    work page Pith review arXiv 2023

  4. [4]

    Escolar, and Shunsuke Tada

    Toshitaka Aoki, Emerson G. Escolar, and Shunsuke Tada. Bipath Persistence.Japan Journal of Industrial and Applied Mathematics, December 2024.doi:10.1007/s13160-024-00681-3

    work page doi:10.1007/s13160-024-00681-3 2024

  5. [5]

    Escolar, Ken Nakashima, and Michio Yoshi- waki

    Hideto Asashiba, Mickaël Buchet, Emerson G. Escolar, Ken Nakashima, and Michio Yoshi- waki. On Interval Decomposability of 2D Persistence Modules.Computational Geometry, 105–106:101879, August 2022.doi:10.1016/j.comgeo.2022.101879

    work page arXiv 2022

  6. [6]

    Approximation by interval-decomposables and interval resolutions of persistence modules

    Hideto Asashiba, Emerson G. Escolar, Ken Nakashima, and Michio Yoshiwaki. Approxi- mation by Interval-Decomposables and Interval Resolutions of Persistence Modules.Jour- nal of Pure and Applied Algebra, 227(10):107397, October 2023. arXiv:2207.03663, doi: 10.1016/j.jpaa.2023.107397

    work page Pith review arXiv 2023

  7. [7]

    Interval Replacements of Persistence Modules, June 2024.arXiv:2403.08308

    Hideto Asashiba, Etienne Gauthier, and Enhao Liu. Interval Replacements of Persistence Modules, June 2024.arXiv:2403.08308. 14 ÁNGEL JA VIER ALONSO AND ENHAO LIU

    arXiv 2024

  8. [8]

    Interval Multiplicities of Persistence Modules, November

    Hideto Asashiba and Enhao Liu. Interval Multiplicities of Persistence Modules, November

    work page

Show all 25 references
  1. [9]

    Generic Multi-Parameter Persistence Modules are Nearly Indecomposable, October 2023.arXiv:2211.15306

    Ulrich Bauer and Luis Scoccola. Generic Multi-Parameter Persistence Modules are Nearly Indecomposable, October 2023.arXiv:2211.15306

    2023 arXiv

  2. [10]

    On the Stability of Interval Decomposable Persistence Mod- ules

    Håvard Bakke Bjerkevik. On the Stability of Interval Decomposable Persistence Mod- ules. Discrete & Computational Geometry , 66(1):92–121, July 2021. doi:10.1007/ s00454-021-00298-0

    2021

  3. [11]

    Algebraic Stability of Zigzag Persistence Modules

    Magnus Botnan and Michael Lesnick. Algebraic Stability of Zigzag Persistence Modules. Algebraic & Geometric Topology, 18(6):3133–3204, October 2018. doi:10.2140/agt.2018. 18.3133

    2018 doi

  4. [12]

    Interval Decomposition of Infinite Zigzag Persistence Modules.Pro- ceedings of the American Mathematical Society, 145(8):3571–3577, January 2017

    Magnus Bakke Botnan. Interval Decomposition of Infinite Zigzag Persistence Modules.Pro- ceedings of the American Mathematical Society, 145(8):3571–3577, January 2017. doi: 10.1090/proc/13465

    2017 doi

  5. [13]

    Signed Barcodes for Multi- ParameterPersistenceviaRankDecompositions.In LIPIcs, Volume 224, SoCG 2022, volume 224, pages 19:1–19:18

    Magnus Bakke Botnan, Steffen Oppermann, and Steve Oudot. Signed Barcodes for Multi- ParameterPersistenceviaRankDecompositions.In LIPIcs, Volume 224, SoCG 2022, volume 224, pages 19:1–19:18. Schloss Dagstuhl – Leibniz-Zentrum für Informatik, 2022.doi:10. 4230/LIPICS.SOCG.2022.19

    2022

  6. [14]

    Magnus Bakke Botnan, Steffen Oppermann, and Steve Oudot. Signed Barcodes for Multi- parameter Persistence via Rank Decompositions and Rank-Exact Resolutions.Foundations of Computational Mathematics, September 2024.doi:10.1007/s10208-024-09672-9

    2024 doi

  7. [15]

    Dan Burghelea and Tamal K. Dey. Topological Persistence for Circle-Valued Maps.Discrete & Computational Geometry, 50(1):69–98, July 2013.doi:10.1007/s00454-013-9497-x

    2013 doi

  8. [16]

    Elder-Rule-Staircodes for Aug- mented Metric Spaces.SIAM Journal on Applied Algebra and Geometry, 5(3):417–454, Jan- uary 2021

    Chen Cai, Woojin Kim, Facundo Memoli, and Yusu Wang. Elder-Rule-Staircodes for Aug- mented Metric Spaces.SIAM Journal on Applied Algebra and Geometry, 5(3):417–454, Jan- uary 2021. doi:10.1137/20M1353605

    2021 doi

  9. [17]

    Dey and Tao Hou

    Tamal K. Dey and Tao Hou. Fast Computation of Zigzag Persistence.LIPIcs, Volume 244, ESA 2022, 244:43:1–43:15, 2022.doi:10.4230/LIPICS.ESA.2022.43

    2022 doi

  10. [18]

    Dey, Woojin Kim, and Facundo Mémoli

    Tamal K. Dey, Woojin Kim, and Facundo Mémoli. Computing Generalized Rank Invariant for 2-Parameter Persistence Modules via Zigzag Persistence and Its Applications.Discrete & Computational Geometry, 71(1):67–94, January 2024.doi:10.1007/s00454-023-00584-z

    2024 doi

  11. [19]

    Isometry Theorem for Continuous Quiver of Type˜A, De- cember 2024

    Xiaowen Gao and Minghui Zhao. Isometry Theorem for Continuous Quiver of Type˜A, De- cember 2024. arXiv:2412.12462, doi:10.48550/arXiv.2412.12462

    2024 doi arXiv

  12. [20]

    Hanson and Job Daisie Rock

    Eric J. Hanson and Job Daisie Rock. Decomposition of Pointwise Finite-Dimensional S1 Persistence Modules.Journal of Algebra and Its Applications, 23(03):2450054, March 2024. doi:10.1142/S0219498824500543

    2024 doi

  13. [21]

    Generalized Persistence Diagrams for Persistence Modules over Posets.Journal of Applied and Computational Topology, 5(4):533–581, December 2021

    Woojin Kim and Facundo Mémoli. Generalized Persistence Diagrams for Persistence Modules over Posets.Journal of Applied and Computational Topology, 5(4):533–581, December 2021. doi:10.1007/s41468-021-00075-1

    2021 doi

  14. [22]

    Interactive Visualization of 2-D Persistence Modules, December 2015

    Michael Lesnick and Matthew Wright. Interactive Visualization of 2-D Persistence Modules, December 2015. arXiv:1512.00180

    2015 arXiv

  15. [23]

    Fock Space Representation of the Circle Quantum Group

    Francesco Sala and Olivier Schiffmann. Fock Space Representation of the Circle Quantum Group. International Mathematics Research Notices, 2021(22):17025–17070, November 2021. doi:10.1093/imrn/rnz268

    2021 doi

  16. [24]

    A Computation of Bipath Persistent Homology and Bipath Persistence Diagrams

    Shunsuke Tada. A Computation of Bipath Persistent Homology and Bipath Persistence Diagrams. Seminar presented at Asia Pacific Seminar on Applied Topology and Geome- try, September 2024. Slides available athttps://shunsuketada1357.github.io/documents/ slide2024-09-20.pdf. Recor...

    2024

  17. [25]

    Multiparameter Persistence Landscapes.Journal of Machine Learning Re- search, 21(61):1–38, 2020

    Oliver Vipond. Multiparameter Persistence Landscapes.Journal of Machine Learning Re- search, 21(61):1–38, 2020. Institute of Geometry, Graz University of Technology, Graz, Austria Email address: alonsohernandez@tugraz.at Department of Mathematics, Kyoto University, Kitashiraka...

    2020

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.