Pith. sign in

REVIEW 3 major objections 3 minor 3 cited by

Higher-order network adaptivity: co-evolution of higher-order structure and spreading dynamics

T0 review · 3 major / 3 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read This paper shows that when a group's breakup rate grows with the number of infected members, epidemic thresholds rise and abrupt discontinuous transitions become continuous, the opposite of what pairwise-like rewiring produces.

desk verdict A promising new model of higher-order adaptivity whose qualitative story likely survives, but Eq. (7) is not supported by the supplied stability analysis. read the letter →

arxiv 2508.15445 v1 pith:2NEPKTS6 submitted 2025-08-21 physics.soc-ph

Longzhao Liu , Hongwei Zheng , Zhihao Han , Xin Wang , Shaoting Tang This is my paper · ORCID
classification physics.soc-ph
keywords higher-orderadaptivityhypergraphsco-evolvingnetworksepidemicspreadingbistabilitydiscontinuousphasetransitionmean-fieldapproximationcusppoints
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 argues that the way groups break apart during an epidemic can change the qualitative shape of an outbreak, not just its final size. Earlier adaptive-network models let every contact involving an infected person rewire at the same rate; the paper calls this pairwise-like adaptivity and shows it widens the region where extinction and outbreak coexist. It then introduces higher-order adaptivity, in which a group's breakup rate grows as r times j to the power h, where j is the number of infected members, and shows this raises the epidemic threshold but shrinks or removes the bistable region and turns discontinuous transitions into continuous ones. The practical upshot is that the same protective instinct, enacted with awareness of how many infected people are inside a group, can have the opposite dynamical effect from blanket avoidance. The claim is supported by an analytical threshold formula for 3-uniform hypergraphs and by simulations on synthetic and real human-contact hypergraphs.

What carries the argument

The load-bearing object is the power-law hyperedge breaking rate π(j) = r j^h. The exponent h is the switch between two regimes: h = 0 dissolves groups at a rate independent of infection count, reproducing pairwise-like adaptivity; h > 0 makes group breakup nonlinearly sensitive to the number of infected members, the defining feature of higher-order adaptivity. The supporting machinery is a hyperedge-based mean-field approximation that tracks l_{m,k}, the number of m-hyperedges containing k infected nodes, assuming the total number of hyperedges E stays constant. Linearizing the resulting ODE system at the disease-free state and applying the Routh–Hurwitz criterion yields the closed-form out

What would settle it

Measure, in a real co-location or contact setting, how a group's breakup probability scales with the number of infected members j: if the scaling exponent is close to zero, the predicted suppression of bistability should disappear. Alternatively, run the same model with a delay before new hyperedges form; as delay grows, hyperedges are lost, and the epidemic threshold should deviate from Eq. (7) while fragmentation-induced abrupt transitions may reappear.

Watch

Extended reading notes

Core claim

The paper claims that higher-order adaptivity—group dissolution that accelerates nonlinearly with the number of infected members—produces qualitatively different epidemic behavior from pairwise-like adaptivity. In the model, an infected person in a group transmits at rate β times j^v, where j is the number of infected members, and a hyperedge containing j infected and at least one susceptible member breaks at rate r times j^h. When h = 0, this reduces to the usual adaptive-rewiring picture, which amplifies or induces bistability and discontinuous transitions. When h > 0, the same adaptivity raises the outbreak threshold but narrows or eliminates the bistable region and changes discontinuous

Load-bearing premise

The derivation keeps the total number of hyperedges fixed by assuming a broken group is immediately replaced by a new group containing a susceptible node; if reformation is delayed, incomplete, or not anchored to susceptible nodes, the threshold formula and the phase-transition predictions no longer follow.

Editorial extensions

If this is right

  • Higher-order adaptivity raises the epidemic threshold, making outbreaks harder to trigger, for both nonlinear higher-order spreading and ordinary pairwise spreading.
  • The bistable region shrinks as h increases and can disappear completely, with cusp points marking where the number and type of phase transitions change.
  • Higher-order adaptivity can also eliminate bistability and discontinuous transitions induced by pairwise-like adaptivity alone, even when the nonlinear infection mechanism is removed (v = 1).
  • The fitness parameter λ does not affect the outbreak threshold but does shrink the bistable region and alter the type of phase transition, meaning better information about who is susceptible changes qualitative behavior without changing onset.
  • The theoretical results are consistent across synthetic uniform hypergraphs and three real human-co-location hypergraphs, so the qualitative contrast is not an artifact of one network structure.

Reading between the lines

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

  • If the assumption of immediate hyperedge reformation is dropped, the total number of hyperedges becomes time-dependent; one would expect fragmentation to reintroduce abrupt extinction and make thresholds deviate from the paper's Eq. (7). This is testable by adding a reformation delay to the same agent-based model.
  • The same nonlinear-breakup rule may apply beyond disease: in opinion cascades, cooperation, or contagion on higher-order structures, a breakup rate that grows with the number of 'active' members could similarly suppress explosive transitions.
  • A direct empirical test would measure how a real group's breakup probability scales with the number of infected members; if the scaling exponent h is near zero in a given setting, the predicted smoothing effect should be absent there.
  • The threshold's independence from λ hints that investing in contact information may not delay the first outbreak, but may change whether the epidemic jumps or grows gradually—an intervention-relevant distinction.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 3 minor

Summary. The paper introduces a co-evolution model of SIS-type spreading on hypergraphs in which hyperedges break at rate π(j)=r j^h and are immediately reformed, keeping the total number of hyperedges constant. For 3-uniform hypergraphs the authors derive a mean-field ODE system and, via linearization and a Routh-Hurwitz analysis, a closed-form outbreak threshold Eq. (7). They report that pairwise-like adaptivity (h=0) enlarges the bistable region and promotes discontinuous transitions, whereas higher-order adaptivity (h>0, together with selective rewiring λ) increases the threshold, shrinks or eliminates bistability, and shifts transitions to continuous. These claims are supported by Gillespie simulations on synthetic and real hypergraphs.

Significance. If correct, the central qualitative finding—that nonlinear (higher-order) dependence of hyperedge breaking on infection level reverses the effects of pairwise-like adaptivity—would be a valuable contribution to adaptive higher-order network theory. The paper offers a tractable mean-field framework, explicit formulas, and extensive simulations on multiple empirical hypergraphs. A notable strength is that the threshold formula contains no fitted parameters. However, the theoretical derivation of Eq. (7) contains a clear sign error in the Routh-Hurwitz conditions, so the quantitative threshold and the phase-diagram lines derived from it are not currently supported. The qualitative simulation results may survive a corrected analysis, but the theoretical framework as written is not reliable.

major comments (3)
  1. [Supplementary Sec. IV.B, Eq. (S9)] The Routh-Hurwitz conditions in Eq. (S8) have reversed signs. For the characteristic polynomial in Eq. (S6), stability of the 3x3 block requires tr<0, det<0, and det/tr - M <0 (equivalently M > det/tr). The paper instead requires the second and third inequalities of Eq. (S8) to be negative, which is the opposite. At β=0, det=(-μ-r)(-2μ-2hr)(-3μ)<0 for μ,r,h>0, so the disease-free state is stable, yet the paper's condition -det<0 would require det>0 and would deem it unstable. Thus Eq. (S9) and Eq. (7) do not follow from the Jacobian (S5). Independently solving det(A)=0 for the Fig. 3(a) parameters (N=3000, E=6000, μ=0.2, r=0.1, v=3, h=0) gives β_c≈0.018, not the value ≈0.016 from Eq. (7). The exact discrepancy may depend on parameters, but the derivation is invalid.
  2. [Supplementary Sec. IV.B, Eq. (S9)] The statement that the 'last formula of Eq. (S8) dominates the outbreak threshold' is not derived; it is asserted through calculation and comparison with numerical solutions of the same ODE system. This is not a substitute for a correct linear stability analysis, especially because the preceding Routh-Hurwitz inequalities are wrong. Since the phase diagrams in Fig. 3 use Eq. (7) as the outbreak threshold, the quantitative theoretical predictions of the paper are unsupported until the threshold is re-derived from the correct stability conditions.
  3. [Model section, constant-E assumption] The model assumes that every broken hyperedge is immediately replaced by a new one, so the total number of hyperedges E is constant. This is explicitly introduced for tractability, and the threshold formula depends on <d>=3E/N being fixed. The central qualitative claims are only tested under this immediate-reformation rule; no robustness analysis is provided for delayed, probabilistic, or incomplete reformation, which would change the co-evolutionary dynamics qualitatively. The title and abstract imply broader applicability, but the demonstrated scope is narrower.
minor comments (3)
  1. [Supplementary Sec. V.A and V.B] 'This session' should be 'This section' in both places.
  2. [Main text, Eq. (7)] The typeset formula is ambiguous: it should be made clear that the denominator applies to the sum Ω + sqrt(...), i.e., β_c = (Ω + sqrt(...)) / (2^{v+2}<d>). The OCR/plain-text rendering currently suggests Ω + [sqrt(...)/denom].
  3. [References] Several references have corrupted characters (e.g., ref. [8] 'D¡ ¯Lima', ref. [10] 'E¡ ª'); the bibliography needs to be cleaned.

Circularity Check

1 steps flagged · score 2.0 of 10

No load-bearing self-citation or parameter fit; sole mild circular element is the selection of the dominant Routh–Hurwitz inequality by comparison with numerical solutions of the same model.

  1. other [Supplementary Materials, Sec. IV.B, just before Eq. (S9)]
    "Through calculation and comparison with numerical solutions of Eq.(S1), we find that the last formula of Eq.(S8) dominates the outbreak threshold, which can be simplified as"

    The closed-form threshold Eq. (S10) is presented as a theoretical prediction, but the step that selects which Routh–Hurwitz inequality is the binding one is justified by comparison with numerical solutions of Eq. (S1) — the very same ODE model whose threshold the formula is then claimed to predict. The paper gives no analytic proof of dominance; it uses the numerically computed threshold to identify the correct inequality and then reports the resulting algebraic expression as a first-principles result. This is not a parameter fit and the functional form is still algebraically determined, so the circularity is mild, but the 'prediction' is partially calibrated to the target numerical solution rather than fully derived from the Jacobian.

full rationale

The central derivation is self-contained: Eq. (7) follows from linearizing the model equations (5)–(6) around the disease-free state and analyzing the 3×3 Jacobian block; no adjustable constant is fitted to simulations, and all parameters (µ, r, h, v, λ, <d>) are hand-set inputs. The claimed qualitative effects (higher-order adaptivity eliminating bistability and shifting transitions from discontinuous to continuous) are supported by direct simulations on synthetic and real hypergraphs and do not depend on the closed-form threshold alone. Citations to the authors' own prior work (refs. [4], [7], [44]) are contextual and not load-bearing; no uniqueness theorem or ansatz is imported from those papers. The only borderline element is in Supplementary Sec. IV.B, where the dominant Routh–Hurwitz inequality is identified 'by comparison with numerical solutions of Eq.(S1)' before being simplified into Eq. (S9)/(S10); this is an in-sample selection rather than a fitted parameter, so it raises the score only slightly. The skeptically noted Routh–Hurwitz sign inconsistency is a mathematical-correctness concern, not a circularity concern, and is not scored here.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The model introduces no new physical entities. The central claims rest on the chosen functional forms of adaptivity (power-law breaking, fitness-based formation) and on the mean-field closure. The parameters r, h, lambda, v, mu are hand-set simulation inputs, not estimated from data. The derivation uses standard linear stability analysis with an unproven dominance step for one Routh-Hurwitz condition.

free parameters (5)
  • r
    Basic hyperedge breaking rate, set by hand (e.g., r=0.1 in Fig. 3) to explore dynamics, not fitted to data.
  • h
    Nonlinearity exponent of the breaking rate, set by hand (h=0,1,4,6,9) to define higher-order adaptivity.
  • lambda
    Susceptible fitness in rewiring, set by hand (lambda=1,60), not fitted to data.
  • v
    Nonlinearity of the infection rate, set by hand (v=1,3), not fitted.
  • mu
    Recovery rate, set to mu=0.2 in all simulations, a standard epidemic model input.
assumptions (6)
  • domain assumption Hyperedge-based mean-field closure: dynamics of hyperedge-type counts l_{m,k} are closed at the level of single hyperedges, ignoring correlations between hyperedges.
    Invoked in Eqs. (5)-(6); the threshold formula and phase diagrams rest on this approximation.
  • ad hoc to paper Constant total number of hyperedges E with immediate reformation of a broken hyperedge.
    Introduced 'for the sake of analytical tractability' in the Model section; not empirically grounded; reduces system dimension and is load-bearing for the threshold formula.
  • domain assumption Power-law functional forms for infection rate theta=beta*j^v and breaking rate pi=r*j^h.
    Assumed based on cited empirical/mechanistic works; no specific data are used to set these forms.
  • domain assumption Fitness-based selection probability Gamma = lambda*Ns/(lambda*Ns + (N-Ns)) with lambda >= 1.
    Inspired by evolutionary game theory (ref. 39); models information quality, not measured empirically.
  • standard math Routh-Hurwitz stability criterion and linearization around the disease-free fixed point.
    Used in SI Sec. IV.B to derive the outbreak threshold.
  • ad hoc to paper The last inequality of Eq. (S8) dominates the stability condition.
    Authors state this is found 'through calculation and comparison with numerical solutions'; the other Routh-Hurwitz conditions are not proven to be automatically satisfied.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Higher-order network adaptivity: co-evolution of higher-order structure and spreading dynamics." pith.science (2026). https://pith.science/paper/2NEPKTS6

@misc{pith2026250815445,
  author       = {Pith},
  title        = {Pith review of: Higher-order network adaptivity: co-evolution of higher-order structure and spreading dynamics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2NEPKTS6}},
  note         = {Machine review of arXiv:2508.15445}
}
read the original abstract

The co-evolution of structure and dynamics, known as adaptivity, is a fundamental property in various systems and drives diverse emergent behaviors. However, the adaptivity in previous works is primarily stemmed from pairwise situations, while is insufficient to capture ubiquitous higher-order characteristics of real systems. Here, we introduce higher-order network adaptivity to model the co-evolution of higher-order structure and spreading dynamics, and theoretically analyze the thresholds and spreading sizes. Results demonstrate that both pairwise-like and higher-order adaptivity consistently increase spreading thresholds, but surprisingly produce completely opposing qualitative effects. Specifically, contrary to pairwise-like adaptivity, higher-order adaptivity not only reduces or even eliminates the bistable region, but also leads to shifts of phase transitions from discontinuous to continuous. These findings are validated on both synthetic and real hypergraphs. Our work introduces an idea of higher-order adaptivity and highlights its fundamental differences from pairwise-like adaptivity, advancing further researches of adaptive higher-order systems.

Figures

Figures reproduced from arXiv: 2508.15445 by the authors.

Figure 1
Figure 1. FIG. 1. Higher-order adaptive spreading model. It describes [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Results based on a real hypergraph derived from Thiers13 data [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Theoretical and simulation results on uniform hypergraphs [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Immunization on Temporal Higher-Order Networks

    physics.soc-ph 2026-07 conditional novelty 6.0 of 10

    On temporal higher-order networks, immunization shows bistability and discontinuous transitions so effectiveness depends on initial prevalence; High Infection Contribution and prevalence-adaptive egocentric strategies...

  2. Higher-order adaptive behaviors outperform pairwise strategies in mitigating contagion dynamics

    physics.soc-ph 2026-02 conditional novelty 6.0 of 10

    On hypergraph SIS models, awareness strategies based on absolute, higher-order information reduce epidemic prevalence more and impose lower social cost than pairwise or relative strategies, by concentrating alert on h...

  3. Adaptive behaviors neutralize bistable explosive transitions in higher-order contagion

    physics.soc-ph 2026-01 conditional novelty 6.0 of 10

    Risk-driven local behavioral adaptation suppresses bistable explosive transitions in hypergraph contagion, and sufficiently strong group-based awareness makes the transition continuous and indistinguishable from pairw...

Reference graph

Works this paper leans on

48 extracted references · 43 canonical work pages · cited by 3 Pith papers

  1. [1]

    Each node has two states: infected ( I) and sus- ceptible (S)

    Firstly, we consider a typical epidemic spreading process on hypergraphs, where hyperedges of size m encode higher-order interactions among m indi- viduals. Each node has two states: infected ( I) and sus- ceptible (S). Infected individuals recover at rate µ , while transmitting the disease to susceptible ones at a certain infection rate. Noteworthy, rece...

    work page 1993

  2. [2]

    Boccaletti, V

    S. Boccaletti, V. Latora, Y. Moreno, M. Chavez, and D.- U. Hwang, Physics reports 424, 175 (2006)

    work page 2006

  3. [3]

    S. N. Dorogovtsev, A. V. Goltsev, and J. F. Mendes, Reviews of Modern Physics 80, 1275 (2008)

    work page 2008

  4. [4]

    Barrat, M

    A. Barrat, M. Barthelemy, and A. Vespignani, Dynami- cal processes on complex networks (Cambridge university press, 2008)

    2008

  5. [5]

    W. Zhu, X. Wang, C. Wang, L. Liu, H. Zheng, and S. Tang, New Journal of Physics 26, 033046 (2024)

    work page 2024

  6. [6]

    Berner, T

    R. Berner, T. Gross, C. Kuehn, J. Kurths, and S. Yanchuk, Physics Reports 1031, 1 (2023)

    work page 2023

  7. [7]

    Gross and B

    T. Gross and B. Blasius, Journal of the Royal Society Interface 5, 259 (2008)

    2008

  8. [8]

    L. Liu, X. Wang, Y. Zheng, W. Fang, S. Tang, and Z. Zheng, New Journal of Physics 22, 013019 (2020)

    work page 2020

Show all 48 references
  1. [9]

    Gross, C

    T. Gross, C. J. D. D¡ ¯Lima, and B. Blasius, Physical review letters 96, 208701 (2006)

    2006

  2. [10]

    X. Wang, A. D. Sirianni, S. Tang, Z. Zheng, and F. Fu, Physical Review X 10, 041042 (2020)

    2020

  3. [11]

    F. Fu, C. Hauert, M. A. Nowak, and L. Wang, Phys- ical Review E¡ ªStatistical, Nonlinear, and Soft Matter Physics 78, 026117 (2008)

    2008

  4. [12]

    Battiston, G

    F. Battiston, G. Cencetti, I. Iacopini, V. Latora, M. Lu- cas, A. Patania, J.-G. Young, and G. Petri, Physics re- ports 874, 1 (2020)

    2020

  5. [13]

    Santoro, F

    A. Santoro, F. Battiston, G. Petri, and E. Amico, Nature Physics 19, 221 (2023)

    2023

  6. [14]

    Grilli, G

    J. Grilli, G. Barab´ as, M. J. Michalska-Smith, and S. Allesina, Nature 548, 210 (2017)

    2017

  7. [15]

    Burgio, S

    G. Burgio, S. G´ omez, and A. Arenas, Physical Review Letters 132, 077401 (2024)

    2024

  8. [16]

    Lambiotte, M

    R. Lambiotte, M. Rosvall, and I. Scholtes, Nature physic s 15, 313 (2019)

    2019

  9. [17]

    Shi and G

    D. Shi and G. Chen, National Science Review 9, nwac038 (2022)

    2022

  10. [18]

    Boccaletti, P

    S. Boccaletti, P. De Lellis, C. Del Genio, K. Alfaro- Bittner, R. Criado, S. Jalan, and M. Romance, Physics Reports 1018, 1 (2023)

    2023

  11. [19]

    Ferraz de Arruda, A

    G. Ferraz de Arruda, A. Aleta, and Y. Moreno, Nature Reviews Physics 6, 468 (2024)

    2024

  12. [20]

    Battiston, E

    F. Battiston, E. Amico, A. Barrat, G. Bianconi, G. Fer- raz de Arruda, B. Franceschiello, I. Iacopini, S. K´ efi, V. Latora, Y. Moreno, et al. , Nature Physics 17, 1093 (2021)

    2021

  13. [21]

    Majhi, M

    S. Majhi, M. Perc, and D. Ghosh, Journal of the Royal Society Interface 19, 20220043 (2022)

    2022

  14. [22]

    Malizia, S

    F. Malizia, S. Lamata-Ot ´ ın, M. Frasca, V. Latora, and J. G´ omez-Garde˜ nes, Nature communications 16, 555 (2025)

    2025

  15. [23]

    Neuh¨ auser, A

    L. Neuh¨ auser, A. Mellor, and R. Lambiotte, Physical Re- view E 101, 032310 (2020)

    2020

  16. [25]

    Civilini, O

    A. Civilini, O. Sadekar, F. Battiston, J. G´ omez- Garde˜ nes, and V. Latora, Physical Review Letters 132, 167401 (2024)

    2024

  17. [26]

    Di Gaetano, F

    L. Di Gaetano, F. Battiston, and M. Starnini, Physical Review Letters 132, 037401 (2024)

    2024

  18. [27]

    Petri and A

    G. Petri and A. Barrat, Physical review letters 121, 228301 (2018)

    2018

  19. [28]

    Z. Han, L. Liu, X. Wang, Y. Hao, H. Zheng, S. Tang, and Z. Zheng, Chaos: An Interdisciplinary Journal of Nonlinear Science 34 (2024)

    2024

  20. [29]

    Iacopini, M

    I. Iacopini, M. Karsai, and A. Barrat, Nature Communi- cations 15, 7391 (2024)

    2024

  21. [30]

    St-Onge, H

    G. St-Onge, H. Sun, A. Allard, L. H´ ebert-Dufresne, and G. Bianconi, Physical review letters 127, 158301 (2021)

    2021

  22. [31]

    W. Wang, Y. Nie, W. Li, T. Lin, M.-S. Shang, S. Su, Y. Tang, Y.-C. Zhang, and G.-Q. Sun, Physics Reports 1056, 1 (2024)

    2024

  23. [32]

    Y. Chen, Y. R. Gel, M. V. Marathe, and H. V. Poor, Proceedings of the National Academy of Sciences 121, e2313171120 (2024)

    2024

  24. [33]

    Marceau, P.-A

    V. Marceau, P.-A. No¨ el, L. H´ ebert-Dufresne, A. Allard, and L. J. Dub´ e, Physical Review E¡ ªStatistical, Nonlin- ear, and Soft Matter Physics 82, 036116 (2010)

    2010

  25. [34]

    Burgio, G

    G. Burgio, G. St-Onge, and L. H´ ebert-Dufresne, Nature Communications 16, 4589 (2025)

    2025

  26. [35]

    S. Lai, N. W. Ruktanonchai, L. Zhou, O. Prosper, W. Luo, J. R. Floyd, A. Wesolowski, M. Santillana, C. Zhang, X. Du, et al. , nature 585, 410 (2020)

    2020

  27. [36]

    Perra, Physics Reports 913, 1 (2021)

    N. Perra, Physics Reports 913, 1 (2021)

    2021

  28. [37]

    Flaxman, S

    S. Flaxman, S. Mishra, A. Gandy, H. J. T. Unwin, T. A. Mellan, H. Coupland, C. Whittaker, H. Zhu, T. Berah, J. W. Eaton, et al. , Nature 584, 257 (2020)

    2020

  29. [38]

    St-Onge, I

    G. St-Onge, I. Iacopini, V. Latora, A. Barrat, G. Petri, A. Allard, and L. H´ ebert-Dufresne, Communications Physics 5, 25 (2022)

    2022

  30. [39]

    Pastor-Satorras, C

    R. Pastor-Satorras, C. Castellano, P. Van Mieghem, and A. Vespignani, Reviews of modern physics 87, 925 (2015)

    2015

  31. [40]

    M. A. Nowak, Evolutionary dynamics: exploring the equations of life (Harvard university press, 2006)

    2006

  32. [44]

    Dhooge, W

    A. Dhooge, W. Govaerts, Y. A. Kuznetsov, H. G. E. Mei- jer, and B. Sautois, Mathematical and Computer Mod- elling of Dynamical Systems 14, 147 (2008)

    2008

  33. [45]

    L. Liu, X. Wang, X. Chen, S. Tang, and Z. Zheng, Fron- tiers in Physics 9, 649852 (2021)

    2021

  34. [46]

    N. E. Friedkin, A. V. Proskurnikov, R. Tempo, and S. E. Parsegov, Science 354, 321 (2016)

    2016

  35. [47]

    Alvarez-Rodriguez, F

    U. Alvarez-Rodriguez, F. Battiston, G. F. de Arruda, Y. Moreno, M. Perc, and V. Latora, Nature Human Be- haviour 5, 586 (2021). Supplementary Materials CONTENTS I. Data and Real Hypergraphs 2 II. Model Simulation 2 III. Results on Two Other Real Hypergraphs 4 IV. Theoretical A...

    2021

  36. [48]

    Mastrandrea, J

    R. Mastrandrea, J. Fournet, and A. Barrat, PloS one 10, e0136497 (2015)

    2015

  37. [49]

    G´ enois and A

    M. G´ enois and A. Barrat, EPJ Data Science 7, 1 (2018)

    2018

  38. [50]

    Iacopini, G

    I. Iacopini, G. Petri, A. Barrat, and V. Latora, Nature communications 10, 2485 (2019)

    2019

  39. [51]

    Young, G

    J.-G. Young, G. Petri, F. Vaccarino, and A. Patania, Physical Review E 96, 032312 (2017)

    2017

  40. [52]

    Dhooge, W

    A. Dhooge, W. Govaerts, Y. A. Kuznetsov, H. G. E. Meijer, and B. Sautois , Mathematical and Computer Modelling of Dynamical Systems 14, 147 (2008). 9

    2008

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.