Pith. sign in

REVIEW 4 major objections 4 minor 37 references

Self-organization to multicriticality

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read An adaptive network of coupled oscillators can self-organize to the simultaneous onset of two distinct phase transitions—a Hopf bifurcation and a Turing bifurcation—using only local plasticity rules, a state the authors call…

desk verdict A genuine first demonstration of self-organization to multicriticality, but the robust-genericity claim overreaches the single-run evidence. read the letter →

arxiv 2506.04275 v2 pith:P6N4OLZT submitted 2025-06-04 nlin.AO

Silja Sormunen , Thilo Gross , Jari Saramäki This is my paper
classification nlin.AO MSC 37G1037N2505C82 PACS 05.65.+b89.75.Hc05.45.-a
keywords self-organizedcriticalitymulticriticalityadaptivenetworksFitzHugh-NagumomodelHopfbifurcationTuringmasterstabilityfunctioncriticalmanifold
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

An adaptive network, whose edge weights change according to local rules, normally self-organizes to one critical point; this paper shows it can go further and self-organize to multicriticality—sitting simultaneously on the onset of two different phase transitions. The system is a network of FitzHugh-Nagumo oscillators that first tunes its leading Laplacian eigenvalue to the Hopf bifurcation (onset of oscillations), then drifts along that critical manifold while the second-smallest Laplacian eigenvalue rises to the Turing bifurcation (onset of pattern formation). The result matters because it overturns the usual picture of self-organized criticality as a single tuned point and suggests that systems like the brain could be poised at several transitions at once, using simultaneous critical states for different functions.

What carries the argument

The load-bearing object is the master stability function decomposition, which writes the eigenvalues of the full-system Jacobian as the union of eigenvalues of reduced matrices P − λ_l C over the Laplacian eigenvalues λ_l of the network. Hopf bifurcation is controlled by the leading Laplacian eigenvalue λ1 through the trace of the reduced matrix, and Turing bifurcation by the second-smallest Laplacian eigenvalue λ_{n-1} through its determinant; the two plasticity rules are designed to target these two eigenvalues separately. The Hopf subcriticality rule weakens strong edges at high-amplitude nodes (nodes with high eigenvector centrality), while the Turing subcriticality rule strengthens weak edges at nodes whose time-averaged state deviates from neighbors (nodes with low eigenvector centrality), and the inequality in Eq. (5) is the time-scale separation that lets λ_{n-1} creep to λ*_T while λ1 is held at λ*_H.

What would settle it

Inactivate the Hopf subcriticality rule while keeping the Turing rule active: if λ_{n-1} still reaches λ*_T while λ1 stays above λ*_H, then the reported drift along the Hopf manifold is not what causes multicriticality. Alternatively, use a regular network where all nodes have equal eigenvector centrality, so both rules update the same types of edges; if the system still reaches multicriticality, then the eigenvector-centrality separation behind Eq. (5) is not necessary.

Watch

Extended reading notes

Core claim

The paper's central claim is that a self-organizing system can become critical in two distinct ways at once, and it presents the first demonstration of this 'self-organized multicriticality.' In a weighted random network of coupled FitzHugh-Nagumo oscillators with two local plasticity rules, the system first self-organizes to the onset of oscillations (Hopf bifurcation), and then, while remaining at that onset, drifts to the onset of pattern formation (Turing bifurcation), so that the leading Laplacian eigenvalue λ1 is pinned at λ*_H while the second-smallest eigenvalue λ_{n-1} rises to λ*_T. The two order parameters—oscillation amplitude and variance across nodes—both hover near zero at the end of the simulation, and frozen-network phase diagrams confirm the system sits at both bifurcation lines. The topology continues to evolve after multicriticality is reached, so the state is a dynamical drift along an intersection of critical manifolds rather than a static point.

Load-bearing premise

The claim rests on the two subcriticality rules affecting the two Laplacian eigenvalues independently enough, so that λ_{n-1} can approach its Turing-critical value while λ1 is held at its Hopf-critical value; this separation is observed in simulations rather than proven, and some network realizations never reach multicriticality.

Editorial extensions

If this is right

  • If the claim holds, self-organized criticality is not limited to a single phase transition: any adaptive system with two bifurcations can reach their intersection and be critical in two ways at once.
  • Because two non-parallel critical hypersurfaces generically intersect, multicritical self-organization should occur in any system with two self-organization rules, not just in this model.
  • For the brain, this predicts that distinct phase transitions (for example, onset of synchrony and onset of pattern formation) can be active at the same time, potentially supporting different computational roles simultaneously.
  • The continuing topology evolution after multicriticality means the critical state is a dynamic manifold, so even at criticality the network structure can keep changing without losing criticality.
  • Equation (5) gives a quantitative design condition: two plasticity mechanisms can be combined into a multicritical self-organizer only if their relative effects on the two controlling eigenvalues satisfy the stated inequality.

Reading between the lines

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

  • A natural extension, not tested in the paper, is a third plasticity rule targeting a third Laplacian eigenvalue; since three non-parallel critical manifolds should meet at a codimension-3 point, tricritical self-organization should be reachable under an analogous separation condition.
  • The eigenvector-centrality separation could serve as an engineering principle: to drive a network to a desired multicritical point, allocate each plasticity rule to the graph region that most strongly shifts its target Laplacian eigenvalue.
  • The phase diagrams in Figs. 2g-h suggest that after reaching multicriticality the transition becomes steeper over time, which we read as the system sharpening its critical sensitivity even while sitting at the double threshold—an interpretation the paper does not make explicitly.
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

4 major / 4 minor

Summary. The paper presents a model of N=50 FitzHugh-Nagumo oscillators coupled on a weighted Erdős–Rényi network whose edge weights evolve according to two local plasticity mechanisms, one targeting the Hopf bifurcation and one the Turing bifurcation. Using master stability function arguments, the authors derive the critical Laplacian eigenvalues λ*_H and λ*_T, and show numerically that the leading eigenvalue λ1 first approaches λ*_H, after which the second-smallest eigenvalue λ_{n−1} increases to λ*_T while λ1 remains at λ*_H. They interpret this as a drift along the Hopf critical manifold to a codimension-2 point where Hopf and Turing instabilities occur simultaneously, and they support the claim with order parameters and frozen-network phase diagrams. The paper concludes that this is the first demonstration of self-organized multicriticality and that such behavior is robust and generic.

Significance. If the claims hold, the paper extends self-organized criticality from a single critical point to a multicritical point where two distinct bifurcations co-occur, providing a new conceptual bridge between adaptive network theory and the critical brain hypothesis. The use of the master stability function to obtain the critical eigenvalues is a genuine strength: the thresholds λ*_H and λ*_T are derived from the model parameters rather than fitted to the simulation, and the order-parameter and phase-diagram checks are appropriate validation tools. The explicit statement of condition (5) as a requirement for drift is also useful, even though it is not quantified. The main significance rests on the generality claim, which as submitted is supported by only a single illustrative trajectory and by references to an unavailable supplementary information.

major comments (4)
  1. [Section III, Fig. 2] The central claim that evolution to multicriticality is 'robust generic behavior' is supported by a single representative run; Fig. 2 shows one trajectory with no ensemble statistics, error bars, or success-rate information. The text itself admits that some network realizations fail to reach multicriticality (SI VI). To make the claim of robustness load-bearing, the authors should report statistics over many independent initial network realizations, including the fraction that reach multicriticality and the distribution of times to reach it.
  2. [Section III, Eq. (5)] Equation (5) is the load-bearing condition for the drift along the Hopf critical manifold, but it is never directly measured. The heuristic explanation based on eigenvector centralities is plausible, yet the authors also state that the condition can fail. The manuscript should quantify the ratios Δλ_{n−1,Tsub}/Δλ_{1,Tsub} and |Δλ_{n−1,Hsub}|/|Δλ_{1,Hsub}| over the course of the shown run and, if possible, across realizations, demonstrating that Eq. (5) holds when drift occurs and is violated when it does not.
  3. [Section III, initialization requirement] The model requires initialization in a supercritical state with respect to both bifurcations, and the supercriticality rules' effects on the eigenvalues are reported to invert during most of the drift (SI IV). This means the system is not self-organizing from generic initial conditions but is placed in a special region of parameter space from which the subcriticality rules push it to multicriticality. The abstract's 'robust generic behavior' and the discussion's 'widespread generic behaviour' should be tempered, or the initialization requirement should be stated explicitly as a scope limitation.
  4. [Sections III and IV (SI references)] The robustness claims, including the dependence on adaptation parameters (SI V), examples of failed realizations (SI VI), and a substantially modified update rule (SI VII), are deferred to a supplementary information that is not included with the submitted manuscript. These appendices are essential for evaluating the generality claim, not merely peripheral. The authors should either provide the SI or reduce the strength of the robustness statements to what can be verified from the main text.
minor comments (4)
  1. [Fig. 2 caption] The caption lists '(a-e) Time evolution of ... and the network average degree ⟨k⟩' but then refers to panel (e) as 'Average eigenvector centrality values'; the main text also calls Fig. 2e the average weighted degree. This inconsistency should be corrected, and panel labels should match the descriptions in both the caption and the body.
  2. [Author affiliations and formatting] The author names contain garbled accented characters (e.g., 'Saram¨aki'), and the institution names are truncated (e.g., 'Es poo'); these should be fixed in the final version.
  3. [Section II.A, Eq. (1)] The term 'adimensionalised' should be 'nondimensionalised' or 'dimensionless'; also, the sentence beginning 'Denoting the adjacency matrix of a network with A' could be rephrased for clarity.
  4. [References] References [24] and [29] have incomplete author lists or formatting artifacts (e.g., 'Current opinion in neurobiology' and page ranges); these should be standardized to the journal style.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the multicritical thresholds are derived from the linearized Jacobian, the adaptive rules are local and threshold-based rather than eigenvalue-targeted, and the self-citations are background rather than load-bearing.

full rationale

The paper's central derivation is self-contained. The critical values λ*_H and λ*_T are obtained from the master stability function analysis of the linearized system (Eq. 3 and surrounding text), not fitted to the simulation or defined by the plasticity rules. The plasticity rules are local: the Hopf subcriticality rule responds to oscillation amplitudes exceeding θH, and the Turing subcriticality rule responds to spatial variation of the time-averaged state exceeding θT; neither rule is given the target eigenvalues. The observed convergence of λ1 and λn−1 to the theoretically computed critical values is therefore an emergent result rather than a tautology. Eq. (5) is a stated condition for the drift to occur, explained heuristically and not derived from the plasticity rules, but it is not used as an input that forces the result; the paper explicitly admits that some network realizations fail to satisfy it and do not reach multicriticality. The self-citations [4,5] provide background and the motivating hypothesis but the confirmation here is a new simulation, so they are not load-bearing. The acknowledged limitations — initialization in a supercritical state and dependence on network realization — are correctness and generality concerns, not circularity. The phase-diagram validation manipulates eigenvalues and measures order parameters after freezing the network; this is a consistency check against the same theoretical thresholds, not a fit. Overall, no step in the derivation reduces by construction to its own input.

Assumptions & free parameters 10 free parameters · 6 assumptions · 0 invented entities

The central claim rests on ten hand-chosen parameter values and six structural assumptions. No new physical entities are introduced. The most important free parameters are the thresholds θH and θT and the ratio β, which controls eigenvalue separation. The most fragile assumption is the unproven independence condition in Eq. (5), together with the required supercritical initialization.

free parameters (10)
  • FitzHugh-Nagumo constants a and b = a=0.8, b=10.5
    Chosen so the reduced determinant as a function of λ is an upward-opening parabola, enabling a Turing bifurcation; hand-selected, not fitted to data.
  • Coupling matrix C = C11=-1.4, C12=0.3, C21=-6.8, C22=0.9
    Hand-chosen so that Hopf and Turing bifurcations can occur independently and simultaneously within the accessible eigenvalue range.
  • Hopf subcriticality threshold θH = 0.05
    Amplitude threshold above which nodes reduce strong links; hand-selected.
  • Turing subcriticality threshold θT = 0.005
    Neighbor-deviation threshold that triggers Turing subcriticality updates; hand-selected.
  • Turing rule strength ratio β = 0.1
    Sets the relative magnitude of Turing versus Hopf subcriticality updates; chosen to provide the time-scale separation in Eq. (5).
  • Weight update amount δw = 0.001
    Step size for supercriticality updates; hand-selected.
  • Supercriticality update schedule c and probability 0.5 = c=10, p=0.5
    Controls the slow supercriticality update rounds; hand-selected.
  • Integration time per update round s = 5000
    Duration of oscillator dynamics between topology updates; hand-selected.
  • Minimum edge weight wmin = 0.0001
    Lower bound on edge weights to prevent collapse; hand-selected.
  • Initial network and noise parameters = N=50, avg degree 8, weights 1 + noise 0.1, U/V init 0.01, noise 0.01
    Initialization chosen to be supercritical with respect to both bifurcations; the paper states this is required.
assumptions (6)
  • standard math Master stability function separation: eigenvalues of the full Jacobian are the union of eigenvalues of P - λ_l C (Eq. 3).
    Taken from Ref. [36]; used to reduce both Hopf and Turing bifurcation conditions to functions of Laplacian eigenvalues.
  • domain assumption FitzHugh-Nagumo with negative and non-biological coupling is an appropriate general model for coupled oscillators.
    The authors explicitly state they are not modeling biological neurons but a general model; this is a domain choice, not a derived result.
  • domain assumption The two plasticity rules implement the generic slow-supercriticality/fast-subcriticality principle of self-organized criticality (Ref. [18]).
    The design of the rules is justified by analogy to established SOC mechanisms; the specific local update rules are ad hoc to this paper.
  • domain assumption The critical values λ*_H and λ*_T remain constant and valid during the adaptive drift.
    The paper uses the homogeneous steady-state bifurcation analysis throughout even while topology evolves; the discussion notes that in less symmetric models the critical values might change.
  • ad hoc to paper The two subcriticality rules affect λ1 and λn−1 independently enough (Eq. 5).
    The authors observe this separation in some network realizations and note that other realizations fail to reach multicriticality; it is not proven for generic networks.
  • ad hoc to paper The system must be initialized in a supercritical state with respect to both bifurcations.
    Stated explicitly: 'our model requires the system to be initialized in a supercritical state with respect to both bifurcations, as it is the subcriticality rules rather than the supercriticality ones that originally drive the system to multicriticality.'

how reviews work

0 comments
Cite this review

Pith. "Pith review of Self-organization to multicriticality." pith.science (2026). https://pith.science/paper/P6N4OLZT

@misc{pith2026250604275,
  author       = {Pith},
  title        = {Pith review of: Self-organization to multicriticality},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P6N4OLZT}},
  note         = {Machine review of arXiv:2506.04275}
}
read the original abstract

Self-organized criticality is a well-established phenomenon, where a system dynamically tunes its structure to operate on the verge of a phase transition. Here, we show that the dynamics inside the self-organized critical state are fundamentally far more versatile than previously recognized, to the extent that a system can self-organize to a new type of phase transition while staying on the verge of another. In this first demonstration of self-organization to multicriticality, we investigate a model of coupled oscillators on a random network, where the network topology evolves in response to the oscillator dynamics. We show that the system first self-organizes to the onset of oscillations, after which it drifts to the onset of pattern formation while still remaining at the onset of oscillations, thus becoming critical in two different ways at once. The observed evolution to multicriticality is robust generic behavior that we expect to be widespread in self-organizing systems. Overall, these results offer a unifying framework for studying systems, such as the brain, where multiple phase transitions may be relevant for proper functioning.

Figures

Figures reproduced from arXiv: 2506.04275 by the authors.

Figure 1
Figure 1. FIG. 1. Snapshots of three neighboring nodes’ timeseries of [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. Self-organized drift to multicriticality. (a-e) Ti [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    P. Bak, C. Tang, and K. Wiesenfeld, Self-organized criti- cality: An explanation of 1/f noise, Physical Review Let- ters 59, 381 (1987)

    work page 1987

  2. [2]

    D. L. Turcotte, Self-organized criticality, Reports on Progress in Physics 62, 1377 (1999)

    work page 1999

  3. [3]

    Mu˜ noz, Colloquium: Criticality and dynamical scal- ing in living systems, Reviews of Modern Physics 90 (2017)

    M. Mu˜ noz, Colloquium: Criticality and dynamical scal- ing in living systems, Reviews of Modern Physics 90 (2017)

    work page 2017

  4. [4]

    Gross, Not one, but many critical states: A dynami- cal systems perspective, Frontiers in Neural Circuits 15, 614268 (2021)

    T. Gross, Not one, but many critical states: A dynami- cal systems perspective, Frontiers in Neural Circuits 15, 614268 (2021)

    work page 2021

  5. [5]

    Sormunen, T

    S. Sormunen, T. Gross, and J. Saram¨ aki, Critical drift in a neuro-inspired adaptive network, Physical Review Letters 130, 188401 (2023)

    work page 2023

  6. [6]

    Chialvo, Emergent complex neural dynamics, Nature Physics 6, 744 (2010)

    D. Chialvo, Emergent complex neural dynamics, Nature Physics 6, 744 (2010)

    work page 2010

  7. [7]

    Herz and H

    A. Herz and H. J, Earthquake cycles and neural rever- berations: collective oscillations in systems with pulse- coupled threshold elements, Physical Review Letters 76, 1222 (1994). 6

    work page 1994

  8. [8]

    B. A. Pearlmutter and C. J. Houghton, A new hypothesis for sleep: Tuning for criticality, Neural Computation 21, 1622 (2009)

    work page 2009

Show all 37 references
  1. [9]

    Hesse and T

    J. Hesse and T. Gross, Self-organized criticality as a fun- damental property of neural systems, Frontiers in Sys- tems Neuroscience 8, 166 (2014)

    2014

  2. [10]

    Rybarsch and S

    M. Rybarsch and S. Bornholdt, Self-organized criticality in neural network models, in Criticality in Neural Sys- tems (Wiley-VCH, Weinheim, 2012) pp. 227–254

    2012

  3. [11]

    N. M. Timme, N. J. Marshall, N. Bennett, M. Ripp, E. Lautzenhiser, and J. M. Beggs, Criticality maximizes complexity in neural tissue, Frontiers in Physiology 7, 425 (2016)

    2016

  4. [12]

    Zeraati, V

    R. Zeraati, V. Priesemann, and A. Levina, Self- organization toward criticality by synaptic plasticity, Frontiers in Physics 9, 619661 (2021)

    2021

  5. [13]

    J. M. Beggs and D. Plenz, Neuronal avalanches in neocor- tical circuits, Journal of Neuroscience 23, 11167 (2003)

    2003

  6. [14]

    Haldeman and J

    C. Haldeman and J. Beggs, Critical branching captures activity in living neural networks and maximizes the number of metastable states, Physical Review Letters 94, 058101 (2005)

    2005

  7. [15]

    Bornholdt and T

    S. Bornholdt and T. R¨ ohl, Self-organized critical neural networks, Physical Review E 67, 066118 (2003)

    2003

  8. [16]

    Meisel and T

    C. Meisel and T. Gross, Adaptive self-organization in a realistic neural network model, Physical Review E 80, 061917 (2009)

    2009

  9. [17]

    Levina, J

    A. Levina, J. Herrmann, and T. Geisel, Dynamical synapses causing self-organized criticality in neural net - works, Nature Physics 3, 857 (2007)

    2007

  10. [18]

    Droste, A

    F. Droste, A. Do, and T. Gross, Analytical investigation of self-organized criticality in neural networks, J. Roy. Soc. Interface 10, 20120558 (2013)

    2013

  11. [19]

    Yaghoubi, T

    M. Yaghoubi, T. de Graaf, J. G. Orlandi, F. Girotto, M. A. Colicos, and J. Davidsen, Neuronal avalanche dy- namics indicates different universality classes in neurona l cultures, Scientific Reports 8, 3417 (2018)

    2018

  12. [20]

    Habibollahi, B

    F. Habibollahi, B. Kagan, A. Burkitt, and C. French, Critical dynamics arise during structured information presentation within embodied in vitro neuronal networks, Nature Communications 14 (2023)

    2023

  13. [21]

    Meisel, A

    C. Meisel, A. Storch, S. Hallmeyer-Elgner, E. Bullmore, and T. Gross, Failure of adaptive self-organized criticali ty during epileptic seizure attacks, PLoS Comput. Biol. 8, e1002312 (2012)

    2012

  14. [22]

    Kitzbichler, M

    M. Kitzbichler, M. Smith, S. Christensen, and E. Bull- more, Broadband criticality of human brain network syn- chronization, PLoS Comput. Biol. 5, e1000314 (2009)

    2009

  15. [23]

    Linkenkaer-Hansen, V

    K. Linkenkaer-Hansen, V. V. Nikouline, J. M. Palva, and R. J. Ilmoniemi, Long-range temporal correlations and scaling behavior in human brain oscillations, Journal of Neuroscience 21, 1370 (2001)

    2001

  16. [24]

    Wilting and V

    J. Wilting and V. Priesemann, 25 years of criticality in neuroscience—established results, open controversies , novel concepts, Current opinion in neurobiology 58, 105 (2014)

    2014

  17. [25]

    Safaeesirat and S

    A. Safaeesirat and S. Moghimi-Araghi, Critical behav- ior at the onset of synchronization in a neuronal model, Physica A Statistical Mechanics and its Applications 587, 10.1016/j.physa.2021.126503 (2022)

    2022

  18. [26]

    F. Y. K. Kossio, S. Goedeke, B. van den Akker, B. Ibarz, and R.-M. Memmesheimer, Growing critical: Self-organized criticality in a developing neural system, Phys. Rev. Lett. 121, 058301 (2018)

    2018

  19. [27]

    Toker, I

    D. Toker, I. Pappas, J. D. Lendner, and F. J., Conscious- ness is supported by near-critical slow cortical electrody - namics, Proceedings of the National Academy of Sciences 119, 10.1073/pnas.2024455119 (2022)

    2022 doi

  20. [28]

    Kanders, T

    K. Kanders, T. Lorimer, and R. Stoop, Avalanche and edge-of-chaos criticality do not necessarily co-occur in neural networks, Chaos 27, 047408 (2017)

    2017

  21. [29]

    Kanders, H

    K. Kanders, H. Lee, N. Hong, Y. Nam, and R. Stoop, Fingerprints of a second order critical line in developing neural networks, Communications Physics 3, 13 (2020)

    2020

  22. [30]

    Fitzhugh, Impulses and physiological states in theo- retical models of nerve membrane, Biophysical Journal 1, 445 (1961)

    R. Fitzhugh, Impulses and physiological states in theo- retical models of nerve membrane, Biophysical Journal 1, 445 (1961)

    1961

  23. [31]

    Nagumo, S

    J. Nagumo, S. Arimoto, and S. Yoshizawa, An active pulse transmission line simulating nerve axon, Proceed- ings of the IRE 50, 2061 (1962)

    1962

  24. [32]

    Gambino, M

    G. Gambino, M. Lombardo, G. Rubino, and M. Sam- martino, Pattern selection in the 2D FitzHugh–Nagumo model, Ricerche di Matematica 68, 535 (2018)

    2018

  25. [33]

    L. A. Segel and S. A. Levin, Application of nonlinear stability theory to the study of the effects of diffusion on predator-prey interactions, AIP Conference Proceedings 27, 123 (1976)

    1976

  26. [34]

    Pecora and T

    L. Pecora and T. Carroll, Master stability functions for synchronized coupled systems, Physical Review Letters 80, 2109 (1998)

    1998

  27. [35]

    Nakao and A

    H. Nakao and A. Mikhailov, Turing patterns in network- organized activator-inhibitor systems, Nature Physics 6, 544 (2010)

    2010

  28. [36]

    Brechtel, P

    A. Brechtel, P. Gramlich, D. Ritterskamp, B. Drossel, and T. Gross, Master stability functions reveal diffusion- driven pattern formation in networks, Physical Review E 97, 032307 (2018)

    2018

  29. [37]

    Abreu, Old and new results on algebraic connectivity of graphs, Linear Algebra and its Applications 423, 53 (2007)

    N. Abreu, Old and new results on algebraic connectivity of graphs, Linear Algebra and its Applications 423, 53 (2007)

    2007

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.