Pith. sign in

REVIEW 2 major objections 5 minor 1 cited by

Any connected threshold graph globally synchronizes in the homogeneous Kuramoto model.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Connected threshold graphs—built by repeatedly adding isolated or universal vertices—are globally synchronizing for the homogeneous Kuramoto model at any edge density.

T0 review reviewed 2026-08-03 challenge →

load-bearing objection Threshold graphs are globally synchronizing — a new class with arbitrary density — but the proof has a fixable gap in the propagation lemma. the 2 major comments →

arxiv 2511.12646 v6 pith:MICMD2OX submitted 2025-11-16 math.DS math.CAmath.COmath.OC

Global synchronization beyond dense graphs: the case of threshold graphs

Hongjin Wu , Ulrik Brandes This is my paper
classification math.DS math.CAmath.COmath.OC MSC 34C1505C7537N2590C26
keywords Kuramoto modelglobal synchronizationthreshold graphssecond-order stationary pointphasor geometryclosed twinsgradient flowsynchronization landscape
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

The paper proves that every connected threshold graph is globally synchronizing: in the homogeneous Kuramoto model with identical frequencies, almost every initial condition converges to a fully synchronized state. Threshold graphs are built by repeatedly adding either a vertex connected to everyone (a dominating vertex) or to no one (an isolated vertex), and they range from star graphs to complete graphs, realizing every possible edge density. The proof shows that at any second-order stationary point of the Kuramoto energy—candidates for non-synchronous stable equilibria—the local phasor geometry forces all oscillators to align. This matters because existing global synchronization guarantees require high minimum degree or expansion, whereas threshold graphs can be arbitrarily sparse while still synchronizing. The mechanism is structural: local symmetries propagate step by step along the construction sequence until the whole graph is locked to one phase.

Core claim

The central claim is Theorem 1.2: any connected threshold graph globally synchronizes. Equivalently, the Kuramoto energy on such a graph has no spurious second-order stationary points: every configuration where the gradient vanishes and the Hessian is positive semidefinite has all phases equal. The proof proceeds by induction on the blocks of dominating vertices in the threshold construction sequence, starting from the last block (which forms a set of closed twins and hence must synchronize) and using the synchronous-pendant lemmas to force each earlier block to adopt the same phase. The argument never needs the half-circle lemma or any density/expansion assumption; it relies entirely on pla

What carries the argument

The argument's engine is a planar phasor-geometric fact (Lemma 5.1): if two unit vectors v_a, v_b satisfy v_b + q = μ_a v_a and v_a + q = μ_b v_b with μ_a, μ_b ≥ 0, then v_a = v_b. This is applied to graph-theoretic pairs called closed twins (vertices with identical closed neighborhoods), which must synchronize at any second-order stationary point (Corollary 5.6), and extended to 'geometric twins' via the synchronous-pendant extension lemmas (Lemmas 5.8 and 5.9), which propagate synchronization from a synchronized block to an adjacent block of dominating vertices. The induction in §6.2 walks this propagation backward along the threshold graph's construction sequence, forcing every block to a

Load-bearing premise

The induction depends on Lemma 5.9, whose proof that a synchronized block plus an adjacent dominating block must share a phase is compressed into a one-sentence appeal to 'geometrically stable twins'; if that propagation step fails—for instance, if the zero-sum case in equation (17) is non-vacuous and not covered—the main theorem collapses.

What would settle it

Find a connected threshold graph and a configuration θ such that ∇E(θ)=0, the Hessian ∇²E(θ) is positive semidefinite, but θ is not fully synchronized. The paper claims no such configuration exists for any connected threshold graph; a direct numerical search for small threshold graphs (e.g., the 8-vertex graph with code 01010101) over critical points would settle it.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Global synchronization is compatible with arbitrary sparsity: threshold graphs exist at every admissible edge density, from stars to complete graphs, so minimum-degree thresholds cannot characterize synchronizing graphs.
  • For any connected threshold graph, gradient descent on the Kuramoto energy from a random starting phase reaches the fully synchronized state with probability one, because no spurious second-order stationary points exist.
  • The energy landscape of the Kuramoto model on threshold graphs is benign in the second-order sense: the only local minima (modulo rotation) are the synchronous states.
  • The proof identifies closed twins and geometric twins as the structural source of synchronization, suggesting a symmetry-based route to global synchronization that complements expansion or density.
  • Threshold graphs' degree sequences are extremal under majorization, so the result shows that extreme degree heterogeneity, not homogeneity, can be compatible with global synchronization.

Where Pith is reading between the lines

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

  • The induction suggests a broader closure principle: any graph obtained by successively adding a vertex whose neighborhood is either everything or nothing relative to the current graph may inherit global synchronization from a base case; threshold graphs are exactly this closure, and the same lemmas might apply to subclasses of cographs or nested-neighborhood graphs.
  • The open problem the authors pose—whether adding a universal vertex to any globally synchronizing graph preserves the property—could likely be attacked with the same phasor-sum machinery, since a universal vertex attaches to a synchronized group and Lemma 5.8-like reasoning applies.
  • If the argument extends to weighted threshold graphs (where twin neighborhoods have equal but nonzero weights), then the result would cover a much larger class of coupling matrices, including those arising in consensus and power-grid models.
  • A numerical experiment on small threshold graphs—enumerating all stationary points and checking the Hessian—would provide a direct computational check of the theorem's strongest form.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 5 minor

Summary. The paper studies the homogeneous Kuramoto model on a graph G, with energy E_G(θ)=½∑ A_ij(1−cos(θ_i−θ_j)). It defines G to be second-order globally synchronizing if every second-order stationary point (SOSP) of E_G is fully synchronized, and invokes Lemma A.1 of [GLPR25] to conclude that this implies global synchronization in the almost-everywhere sense. The main theorem (Theorem 1.2) states that every connected threshold graph is second-order globally synchronizing. The proof recasts equilibria as phasor-sum conditions, introduces closed twins and geometric twins, proves local propagation lemmas (Lemmas 5.2, 5.8, 5.9), and then proves Theorem 1.2 by induction on the 1-blocks of the threshold construction sequence. The paper also shows that threshold graphs realize every admissible edge density and that their degree sequences are extremal under majorization.

Significance. If the proof is completed, the result is a significant advance: it provides a large, recursively defined class of graphs that globally synchronize despite having arbitrary edge density and, at the sparse end, minimum degree 1. This demonstrates a purely structural synchronization mechanism — twin symmetries and induction along nested neighborhoods — rather than density or expansion. The proof is self-contained, free of fitted parameters, and the geometric primitives are clearly formulated. The result also gives natural counterexamples to the intuition that sparse global synchronization requires random or expander structure.

major comments (2)
  1. [Lemma 5.9, Step B (Eq. (16))] The proof asserts μ_i+|S_1|−1>0. This is false when |S_1|=1 and μ_i=0. The possibility μ_i=0 is allowed by Lemma 5.8 (the sum over S_2⊎P may be zero) and is not excluded by second-order stationarity: the Hessian argument in Lemma 5.8 only excludes the coefficient being strictly negative, not equal to zero. Consequently, (16) may be a zero vector, and the subsequent use of Corollary 5.2 is not justified. Since Theorem 1.2's induction in §6.2 invokes Lemma 5.9 at every step, this is a load-bearing gap. The gap is patchable (e.g., by treating the zero-sum case separately and showing, via the equilibrium condition for j∈S_2, that v_j=v_i), but the text as written is incomplete.
  2. [Lemma 5.9, Step C (p. 17)] The sentence 'Combining (16) and (17), we deduce that all nodes in S_1⊎S_2 are geometrically stable twins' is non-constructive. To invoke Corollary 5.2 for a pair (i,j)∈S_1×S_2, one must produce a vector q satisfying the two equations of Lemma 5.1 with nonnegative multipliers. The natural candidate q=∑_{k∈P⊎(S_1\{i})⊎(S_2\{j})} v_k requires a derivation, and the zero-sum branch of (17) must be handled explicitly. Neither is supplied. This is the same load-bearing step as the previous comment and needs to be rewritten.
minor comments (5)
  1. [§4.2, Remark 4.2] The remark writes μ_i = |∑_{j∈N(i)} v_j|, but for a general equilibrium μ_i in Lemma 4.1 can be negative; the equality is only valid at a second-order stationary point or with an orientation convention. Please clarify.
  2. [§1.2] There is a typo: '/suppress Lojasiewicz' should be 'Lojasiewicz' (stray slash).
  3. [§5.2, Theorem 5.7] In Case 2 of the proof, 'By an argument analogous to the one above' leaves the Hessian computation implicit. A few lines would make the proof self-contained.
  4. [§6.2, Base case] The claim that all nodes in a block of 1's form closed twins is correct, but it would help to spell out that the later dominating vertices connect back to the earlier ones, so the closed neighborhoods coincide.
  5. [§5.3, Lemma 5.8, Eq. (14)] In the Hessian computation, the contribution from vertices in Q is zero because N(i)⊆S; this is implicit but should be stated for clarity.

Circularity Check

0 steps flagged

No circularity: proof is self-contained, external anchors are independent, and the terse Step C is an omitted verification rather than a circular step.

full rationale

We find no circular step. Theorem 1.2 is proved by a forward derivation from standard definitions: the threshold-graph construction sequence, the geometric SOSP condition (Lemma 4.4), the closed-twin synchronization lemma (Corollary 5.6), and the propagation Lemma 5.9. The external anchors are [GLPR25, Lemma A.1] for the SOSP-implies-global-synchronization implication, [MP05] for the geometric equilibrium rephrasing, and [PS89] for the degree-sequence extremality; none are self-citations and none are fitted to the main result. No parameter is fitted and no prediction is renamed as an input. The one potentially load-bearing inference, Lemma 5.9 Step C, is terse: it asserts without explicitly constructing the common vector q that (16) and (17) make S1⊎S2 geometrically stable twins. That is an omitted verification, not a circular reduction: the zero-sum branch in (17) can be absorbed with μ_j = 0, and the required q is determined by the common sum in (16)–(17). Thus the central derivation does not reduce to its own inputs, and the paper's own equations do not manufacture the conclusion.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The central claim rests on standard Kuramoto energy landscape facts (external implication lemma from GLPR25), the recursive characterization of threshold graphs (Chvátal-Hammer), and the paper's own geometric lemmas. There are no fitted parameters or invented physical entities.

axioms (3)
  • domain assumption Absence of spurious second-order stationary points implies global synchronization (except measure-zero initial conditions)
    Invoked in §1.2 via Lemma A.1 of [GLPR25]; the paper does not prove it, and the main theorem depends on it.
  • domain assumption Threshold graphs are exactly the graphs generated by iteratively adding isolated or dominating vertices
    Standard characterization cited in §2.1; the induction in §6.2 uses this recursive structure as the backbone.
  • standard math All trajectories of the homogeneous Kuramoto gradient flow converge to the set of equilibria (LaSalle invariance)
    Used in §1.2 to frame the dynamics; the formal implication lemma from GLPR25 already subsumes the convergence needed for the main conclusion.

reviewed 2026-08-03 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Global synchronization beyond dense graphs: the case of threshold graphs." pith.science (2026). https://pith.science/paper/MICMD2OX

@misc{pith2026251112646,
  author       = {Pith},
  title        = {Pith review of: Global synchronization beyond dense graphs: the case of threshold graphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MICMD2OX}},
  note         = {Machine review of arXiv:2511.12646}
}
Share X Bluesky LinkedIn Reddit HN
abstract

Given a graph \(G\) with adjacency matrix \(A\), consider the homogeneous Kuramoto energy $E_G(\boldsymbol{\theta}):=\frac{1}{2}\sum_{1\leq i,j\leq n}A_{ij}\bigl(1-\cos(\theta_i-\theta_j)\bigr)$. We call \(G\) \emph{second-order globally synchronizing} if every second-order stationary point of \(E_G\) is fully synchronized. This property implies \emph{global synchronization}, namely that, up to a measure-zero set of initial conditions, trajectories of the Kuramoto model converge to a fully synchronized state. A fundamental graph-theoretic question is to identify which graph structures have this property. Existing guarantees for global synchronization typically require large minimum degree which forces the graph to be very dense, or good expansion properties. In this paper, we show that synchronization can also arise from a different, purely structural mechanism. More precisely, we prove that threshold graphs, a classical recursively defined graph class, are second-order globally synchronizing, and hence globally synchronizing. Thus, globally synchronizing graphs need not be very dense, have large minimum degree, or satisfy strong expansion-type conditions. The proof exploits the recursive construction of threshold graphs: local phasor constraints imposed by second-order stationarity are propagated along the construction sequence until full synchronization is forced.

Figures

Figures reproduced from arXiv: 2511.12646 by Hongjin Wu, Ulrik Brandes.

Figure 1
Figure 1. Figure 1: A spring analogy for a coupled oscillator network. The label on each curved arrow marks which node exerts the force. From the above illustration of the complete graph K3, one may ask whether the three oscillators will eventually converge to a common position when placed as shown in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Four connected threshold graphs on 19 vertices, from the sparsest (star graph) to the densest (complete graph). Node labels indicate the vertex addition order in the construc￾tion, with vertex 1 as the initial vertex. White and black circles denote isolated and dominating vertices (bit 0 and bit 1), and the initial vertex 1 is colored in white. Codes from left to right: 000000000000000001, 0001000000000100… view at source ↗
Figure 3
Figure 3. Figure 3: Forbidden induced subgraphs for threshold graphs. Remark 2.3. Interestingly, threshold graphs appeared in the synchronization literature long before their role in the Kuramoto model was considered. In particular, as early as 1977, Ausiello, Messina, and Protasi [AMP77] introduced a class of graphs (later recognized as threshold graphs) in the study of synchronization primitives in distributed computation. … view at source ↗
Figure 4
Figure 4. Figure 4: Relation between different concepts [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Representation of vi with angle θi . 4.2. Alignment and Equilibrium. Given a vector-labeled graph (G, {vi}i∈V ) associated with the state θ, we reformulate the equilibrium condition (3) (equivalently (4)) in geometric terms. Lemma 4.1 (Equilibrium condition rephrased geometrically). A state θ is an equilibrium of the Kuramoto model (1) on G = (V, E) with adjacency A if and only if, for each i ∈ V , X j∈N(i… view at source ↗
Figure 6
Figure 6. Figure 6: The solid arrow represents vi P , and the dashed arrow represents j∈N(i) vj . For each node i, the four panels show all possible relations between vi and P j∈N(i) vj . Among them, case (A) violates the first-order condition, while case (B) violates the second-order stationary condition, since the aggregated phasor P j∈N(i) vj lies outside the feasible (gray) region corresponding to µi ≥ 0. The geometric co… view at source ↗
Figure 7
Figure 7. Figure 7: The feasible region for (µa, µb) consists of: (1) the black line µa = µb, excluding the intersection point, corresponding to va = vb; (2) the red line µa+µb = −2, excluding the intersection point, corresponding to the antipodal case va = −vb; (3) their intersection point (−1, −1), marked by a hollow dot, at which no additional constraint is imposed on the vectors va and vb. Corollary 5.2. Under the assumpt… view at source ↗
Figure 8
Figure 8. Figure 8: The two black nodes form a pair of closed twins. Open twins, corresponding to N(i) = N(j), will not be needed in this work [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Windmill graphs W5,6. 5.3. Synchronous Pendant Extension. In this section, we present Lemma 5.9 which describes a scenario where a group of nodes forms geometrically stable closed twins 5.4 without being closed twins in the graph-theoretic sense. Before that, we establish Lemma 5.8, which serves as a prepara￾tory step. Lemma 5.8. Let G = (V, E) be a graph, and let W ⊆ V admit a partition W = Q ⊎ S ⊎ P, wit… view at source ↗
Figure 10
Figure 10. Figure 10: Illustration of Lemma 5.8. The green vertices correspond to nodes for which µivi = P j∈N(i) vj holds with µi ≥ 0. The large circular node represents the set P of vertices. Gray-shaded edges between two nodes indicate that the vectors corresponding to these two nodes are synchronized. The structural relations among the sets Q, S, and P, together with the synchronization of Q ⊎ S, imply a synchro￾nization i… view at source ↗
Figure 11
Figure 11. Figure 11: Illustration of the proof of Lemma 5.9, illustrating how the synchro￾nized block Q⊎S1 extends to include S2 through propagation enabled by the specific local structure. The green vertices correspond to nodes for which µivi = P j∈N(i) vj holds with µi ≥ 0. The gray-shaded nodes are synchronized; an edge with a gray background indicates synchronization between the corresponding nodes. Each or￾ange arc separ… view at source ↗
Figure 12
Figure 12. Figure 12: Synchronization propagates in eight steps, forcing every second-order stationary point to be synchronous. Gray-shaded edges represent synchronizing relations that have been established at each step. Step (1) We begin with the edge between H and I. Applying Lemma 4.4 to node H shows that, since θ is a second-order stationary point, we must have vH ⇑ vI . Equivalently, θH = θI . Step (2) Observe that N(G) =… view at source ↗

discussion (0)

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

Forward citations

Cited by 1 Pith paper

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

  1. Benign nonconvexity of synchronization landscape induced by graph skeletons

    math.CO 2025-12 unverdicted novelty 6.0

    Quasi-threshold graphs induce benign nonconvexity in the Kuramoto energy landscape via sequential local synchronization propagating along their skeletons.

Reference graph

Works this paper leans on

54 extracted references · 7 linked inside Pith · cited by 1 Pith paper

  1. [1]

    Guarantees for spontaneous synchronization on random geometric graphs

    Pedro Abdalla, Afonso S Bandeira, and Clara Invernizzi. Guarantees for spontaneous synchronization on random geometric graphs. SIAM Journal on Applied Dynamical Systems , 23(1):779--790, 2024

    2024

  2. [2]

    Bandeira, Martin Kassabov, Victor Souza, Steven H

    Pedro Abdalla, Afonso S. Bandeira, Martin Kassabov, Victor Souza, Steven H. Strogatz, and Alex Townsend. Expander graphs are globally synchronizing, 2024

    2024

  3. [3]

    The kuramoto model: A simple paradigm for synchronization phenomena

    Juan A Acebr \'o n, L L Bonilla, Conrad J P \'e rez Vicente, F \'e lix Ritort, and Renato Spigler. The kuramoto model: A simple paradigm for synchronization phenomena. Reviews of Modern Physics , 77(1):137--185, 2005

    2005

  4. [4]

    Absil and K

    P.-A. Absil and K. Kurdyka. On the stable equilibrium points of gradient systems. Systems & Control Letters , 55(7):573--577, 2006

    2006

  5. [5]

    A graph-theoretic characterization of the pv-chunk class of synchronizing primitives

    Giorgio Ausiello, Alberto Messina, and Giuseppe Protasi. A graph-theoretic characterization of the pv-chunk class of synchronizing primitives. Information and Control , 34(3):155--170, 1977

    1977

  6. [6]

    Mechanism of rhythmic synchronous flashing of fireflies

    John Buck and Elisabeth Buck. Mechanism of rhythmic synchronous flashing of fireflies. Science , 159(3821):1319--1327, 1968

    1968

  7. [7]

    On the low-rank approach for semidefinite programs arising in synchronization and community detection

    Afonso S Bandeira, Nicolas Boumal, and Vladislav Voroninski. On the low-rank approach for semidefinite programs arising in synchronization and community detection. In Conference on learning theory , pages 361--382. PMLR, 2016

    2016

  8. [8]

    Understanding the dynamics of biological and neural oscillator networks through exact mean-field reductions: a review

    Christian Bick, Marc Goodfellow, Carlo R Laing, and Erik A Martens. Understanding the dynamics of biological and neural oscillator networks through exact mean-field reductions: a review. Journal of Mathematical Neuroscience , 10(1):9, 2020

    2020

  9. [9]

    Positive Definite Matrices

    Rajendra Bhatia. Positive Definite Matrices . Princeton University Press, Princeton, NJ, 2007

    2007

  10. [10]

    Bates, Jonathan D

    Daniel J. Bates, Jonathan D. Hauenstein, Andrew J. Sommese, and Charles W. Wampler. Bertini: Software for Numerical Algebraic Geometry . Available at bertini.nd.edu, 2013

    2013

  11. [11]

    Bandeira, Anastasia Kireeva, Antoine Maillard, and Almut RÃdder

    Afonso S. Bandeira, Anastasia Kireeva, Antoine Maillard, and Almut RÃdder. Randomstrasse101: Open problems of 2024, 2025

    2024

  12. [12]

    Eduardo A. Canale. From weighted to unweighted graphs in synchronizing graph theory. arXiv:2209.06362 [math.CO] , 2022

    Pith/arXiv arXiv 2022

  13. [13]

    Chiang, C

    H.-D. Chiang, C. C. Chu, and G. Cauley. Direct stability analysis of electric power systems using energy functions: Theory, applications, and perspective. Proceedings of the IEEE , 83(11):1497--1529, 1995

    1995

  14. [14]

    Chv \'a tal and P

    V. Chv \'a tal and P. L. Hammer. Set-packing problems and threshold graphs. Cahiers du C.E.R.O. , 15(2--3):109--127, 1973

    1973

  15. [15]

    Chv \'a tal and P

    V. Chv \'a tal and P. L. Hammer. Aggregation of inequalities in integer programming. In P. L. Hammer, E. L. Johnson, B. H. Korte, and G. L. Nemhauser, editors, Studies in Integer Programming , volume 1 of Annals of Discrete Mathematics , pages 145--162. North-Holland, New York, 1977

    1977

  16. [16]

    Quantitative clustering in mean-field transformer models

    Shi Chen, Zhengjiang Lin, Yury Polyanskiy, and Philippe Rigollet. Quantitative clustering in mean-field transformer models. arXiv preprint arXiv:2504.14697 , 2025

    Pith/arXiv arXiv 2025

  17. [17]

    McRae, and Nicolas Boumal

    Christopher Criscitiello, Quentin Rebjock, Andrew D. McRae, and Nicolas Boumal. Synchronization on circles and spheres with nonlinear interactions. arXiv preprint arXiv:2405.18273 , 2024

    arXiv 2024

  18. [18]

    Synchronization and transient stability in power networks and non-uniform kuramoto oscillators

    Florian D \"o rfler and Francesco Bullo. Synchronization and transient stability in power networks and non-uniform kuramoto oscillators. SIAM Journal on Control and Optimization , 50(3):1616--1642, 2012

    2012

  19. [19]

    Synchronization in complex networks of phase oscillators: A survey

    Florian D \"o rfler and Francesco Bullo. Synchronization in complex networks of phase oscillators: A survey. Automatica , 50(6):1539--1564, 2014

    2014

  20. [20]

    Dynamic metastability in the self-attention model

    Borjan Geshkovski, Hugo Koubbi, Yury Polyanskiy, and Philippe Rigollet. Dynamic metastability in the self-attention model. arXiv preprint arXiv:2410.06833 , 2024

    Pith/arXiv arXiv 2024

  21. [21]

    The emergence of clusters in self-attention dynamics

    Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet. The emergence of clusters in self-attention dynamics. Advances in Neural Information Processing Systems , 36:57026--57037, 2023

    2023

  22. [22]

    A mathematical perspective on transformers

    Borjan Geshkovski, Cyril Letrouit, Yury Polyanskiy, and Philippe Rigollet. A mathematical perspective on transformers. Bulletin of the American Mathematical Society , 62(3):427--479, 2025

    2025

  23. [23]

    Horn and Charles R

    Roger A. Horn and Charles R. Johnson. Matrix Analysis . Cambridge University Press, Cambridge, UK, 1st edition, 1985

    1985

  24. [24]

    The random graph process is globally synchronizing

    Vishesh Jain, Clayton Mizgerd, and Mehtaab Sawhney. The random graph process is globally synchronizing. Bulletin of the London Mathematical Society , 2025

    2025

  25. [25]

    Hassan K. Khalil. Nonlinear Systems . Prentice Hall, Upper Saddle River, NJ, 3rd edition, 2002

    2002

  26. [26]

    Clustering in causal attention masking

    Nikita Karagodin, Yury Polyanskiy, and Philippe Rigollet. Clustering in causal attention masking. Advances in Neural Information Processing Systems , 37:115652--115681, 2024

    2024

  27. [27]

    Strogatz, and Alex Townsend

    Martin Kassabov, Steven H. Strogatz, and Alex Townsend. Sufficiently dense kuramoto networks are globally synchronizing. Chaos: An Interdisciplinary Journal of Nonlinear Science , 31(7):073135, 2021

    2021

  28. [28]

    A global synchronization theorem for oscillators on a random graph

    Martin Kassabov, Steven H Strogatz, and Alex Townsend. A global synchronization theorem for oscillators on a random graph. Chaos: An Interdisciplinary Journal of Nonlinear Science , 32(9), 2022

    2022

  29. [29]

    Self-entrainment of a population of coupled non-linear oscillators

    Yoshiki Kuramoto. Self-entrainment of a population of coupled non-linear oscillators. In International Symposium on Mathematical Problems in Theoretical Physics , pages 420--422. Springer, 1975

    1975

  30. [30]

    Pointwise convergence of gradient-like systems

    Christian Lageman. Pointwise convergence of gradient-like systems. Mathematische Nachrichten , 280(13--14):1543--1558, 2007

    2007

  31. [31]

    Local geometry determines global landscape in low-rank factorization for synchronization

    Shuyang Ling. Local geometry determines global landscape in low-rank factorization for synchronization. arXiv preprint arXiv:2311.18670 , 2023

    Pith/arXiv arXiv 2023

  32. [32]

    Synchronization of kuramoto oscillators in dense networks

    Jianfeng Lu and Stefan Steinerberger. Synchronization of kuramoto oscillators in dense networks. Nonlinearity , 33(11):5905--5918, 2020

    2020

  33. [33]

    Lu and S

    Y. Lu and S. Steinerberger. Synchronization of the kuramoto model in a high-dimensional network. SIAM Journal on Applied Dynamical Systems , 20(4):2327--2345, 2021

    2021

  34. [34]

    On the landscape of synchronization networks: A perspective from nonconvex optimization

    Shuyang Ling, Ruitu Xu, and Afonso S Bandeira. On the landscape of synchronization networks: A perspective from nonconvex optimization. SIAM Journal on Optimization , 29(3):1879--1907, 2019

    1907

  35. [35]

    Benign landscapes for synchronization on spheres via normalized laplacian matrices

    Andrew D McRae. Benign landscapes for synchronization on spheres via normalized laplacian matrices. arXiv preprint arXiv:2503.18801 , 2025

    Pith/arXiv arXiv 2025

  36. [36]

    Algebraic geometrization of the kuramoto model: Equilibria and stability analysis

    Dhagash Mehta, Noah S Daleo, Florian D \"o rfler, and Jonathan D Hauenstein. Algebraic geometrization of the kuramoto model: Equilibria and stability analysis. Chaos: An Interdisciplinary Journal of Nonlinear Science , 25(5), 2015

    2015

  37. [37]

    D. C. Michaels, E. P. Matyas, and J. Jalife. Mechanisms of sinoatrial pacemaker synchronization: a new hypothesis. Circulation Research , 61(5):704--714, 1987

    1987

  38. [38]

    N. V. R. Mahadev and U. N. Peled. Threshold Graphs and Related Topics , volume 56 of Annals of Discrete Mathematics . Elsevier Science Publishers, 1995

    1995

  39. [39]

    Global considerations on the kuramoto model of sinusoidally coupled oscillators

    Pablo Monz \'o n and Fernando Paganini. Global considerations on the kuramoto model of sinusoidally coupled oscillators. In Proceedings of the 44th IEEE Conference on Decision and Control , pages 3923--3928. IEEE, 2005

    2005

  40. [40]

    Chaotic attractor in the kuramoto model

    Yuri L Maistrenko, Oleksandr V Popovych, and Peter A Tass. Chaotic attractor in the kuramoto model. International Journal of Bifurcation and Chaos , 15(11):3457--3466, 2005

    2005

  41. [41]

    Mikhailov and Kenneth Showalter

    Alexander S. Mikhailov and Kenneth Showalter. Control of waves, patterns and turbulence in chemical and biological systems. Physics Reports , 425(2-3):79--194, 2006

    2006

  42. [42]

    Nabi and Jeff Moehlis

    A. Nabi and Jeff Moehlis. Single input optimal control for globally coupled neuron networks. Journal of Neural Engineering , 8(6):065008, 2011

    2011

  43. [43]

    Flocking for multi-agent dynamic systems: Algorithms and theory

    Reza Olfati-Saber. Flocking for multi-agent dynamic systems: Algorithms and theory. IEEE Transactions on Automatic Control , 51(3):401--420, 2006

    2006

  44. [44]

    Dynamics of globally coupled oscillators: Progress and perspectives

    Arkady Pikovsky and Michael Rosenblum. Dynamics of globally coupled oscillators: Progress and perspectives. Chaos , 25(9):097616, 2015

    2015

  45. [45]

    Synchronization of mean-field models on the circle

    Yury Polyanskiy, Philippe Rigollet, and Andrew Yao. Synchronization of mean-field models on the circle. arXiv preprint arXiv:2507.22857 , 2025

    Pith/arXiv arXiv 2025

  46. [46]

    The mean-field dynamics of transformers

    Philippe Rigollet. The mean-field dynamics of transformers. arXiv preprint arXiv:2512.01868 , 2025

    arXiv 2025

  47. [47]

    Strogatz and Ian Stewart

    Steven H. Strogatz and Ian Stewart. Coupled oscillators and biological synchronization. Scientific American , 269(6):102--109, 1993

    1993

  48. [48]

    Strogatz

    Steven H. Strogatz. From kuramoto to crawford: exploring the onset of synchronization in populations of coupled oscillators. Physica D , 143(1--4):1--20, 2000

    2000

  49. [49]

    Peter A. Tass. A model of desynchronizing deep brain stimulation with a demand-controlled coordinated reset of neural subpopulations. Biological Cybernetics , 89(2):81--88, 2003

    2003

  50. [50]

    There is no non-zero stable fixed point for dense networks in the homogeneous kuramoto model

    Richard Taylor. There is no non-zero stable fixed point for dense networks in the homogeneous kuramoto model. Journal of Physics A: Mathematical and Theoretical , 45(5):055102, 2012

    2012

  51. [51]

    Conditions for the existence of fixed points in a finite system of kuramoto oscillators

    Mark Verwoerd and Oliver Mason. Conditions for the existence of fixed points in a finite system of kuramoto oscillators. In Proceedings of the 2007 American Control Conference , pages 4613--4618. IEEE, 2007

    2007

  52. [52]

    Artificial kuramoto oscillatory neurons

    Huanchen Wang, Tian Qi Chen, Jonathan Pilault, Yoshua Bengio, and Blake Richards. Artificial kuramoto oscillatory neurons. arXiv preprint arXiv:2410.13821 , 2024

    Pith/arXiv arXiv 2024

  53. [53]

    Arthur T. Winfree. Biological rhythms and the behavior of populations of coupled oscillators. Journal of Theoretical Biology , 16(1):15--42, 1967

    1967

  54. [54]

    The lower bound of the network connectivity guaranteeing in-phase synchronization

    Ryosuke Yoneda, Tsuyoshi Tatsukawa, and Jun-nosuke Teramae. The lower bound of the network connectivity guaranteeing in-phase synchronization. Chaos: An Interdisciplinary Journal of Nonlinear Science , 31(6):063124, 2021

    2021

This paper was first reviewed by deepseek-v4-flash on August 3, 2026.