Pith. sign in

REVIEW 3 major objections 5 minor 65 references

Random walks with resetting on hypergraph

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read Resetting a hypergraph random walk is solved in closed form from the spectrum of the reset-free walk.

desk verdict Useful spectral formulas for resetting walks on hypergraphs, but the headline optimal-reset condition is proved only for r=i and there is an extra factor N in the Section V algebra; both need fixing before the paper can be trusted as written. read the letter →

arxiv 2505.04215 v1 pith:S2KEK3ES submitted 2025-05-07 cs.SI physics.data-an

classification cs.SIphysics.data-an MSC 05C6505C8260J10
keywords hypergraphrandomwalksstochasticresettingspectraltheoryoccupationprobabilitystationarydistributionmeanfirstpassagetimenoderankingcover
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

The paper considers a random walk on a hypergraph that, at each step, moves to a neighbour with probability $1-\gamma$ or jumps back to a fixed reset node with probability $\gamma$, and it asks what this resetting does to standard search statistics. Its answer is that the reset process is a rank-one perturbation of the ordinary hypergraph walk: the reset transition matrix $\Pi(r,\gamma)=(1-\gamma)W+\gamma\Theta(r)$ has eigenvalues $1$ and $(1-\gamma)\lambda_l$, and its eigenvectors are explicit linear corrections to those of $W$. From this, the paper obtains closed-form expressions for the occupation probability, the stationary distribution, and the mean first passage time in terms of the reset-free eigenvalues and eigenvectors. It also derives a condition for the optimal reset probability, expressed through the coefficient of variation of the first-passage time, and a sufficient condition for a small reset rate to help. Because the framework keeps the hypergraph intact instead of converting it to a clique graph, node rankings differ from the standard expansion, and low reset rates can shorten cover time.

What carries the argument

The engine of the paper is the rank-one reset matrix $\Theta(r)$, the matrix whose $r$-th column is all ones and whose other entries are zero, added to the plain hypergraph transition matrix as $\Pi(r,\gamma)=(1-\gamma)W+\gamma\Theta(r)$. Because this is a rank-one perturbation, the spectrum of $\Pi$ can be written directly from the spectrum of $W$: non-unit eigenvalues are scaled by $(1-\gamma)$, and each eigenvector is shifted along the all-ones direction $|\phi_1\rangle$ by a coefficient $\gamma\langle r|\phi_l\rangle/\big((1-(1-\gamma)\lambda_l)\langle r|\phi_1\rangle\big)$. That relation is what turns occupation probability, stationary distribution, and mean first passage time into finite sums over the reset-free eigen data, and it is also what converts the optimal-reset problem into the coefficient-of-variation condition $z^2(\gamma^*)=1+1/\langle T(\gamma^*)\rangle$.

What would settle it

Take a small connected hypergraph (for instance three nodes in a single three-node hyperedge), set the reset node $r$ different from the starting node $i$, and compute $\langle T_{ij}(r,\gamma)\rangle$ exactly by solving the $n\times n$ linear system for many $\gamma$. If the minimizer does not satisfy $z^2(\gamma^*)=1+1/\langle T_{ij}(r,\gamma^*)\rangle$, the condition as stated is false; if it holds across all such small cases, that supports extending the proof to $r\neq i$.

Watch

Extended reading notes

Core claim

The central claim is a spectral dictionary between the hypergraph walk with resetting and the one without. Given the spectral decomposition $W=\sum_{l=1}^n \lambda_l|\phi_l\rangle\langle\bar\phi_l|$, with $\lambda_1=1$, the reset matrix $\Pi(r,\gamma)=(1-\gamma)W+\gamma\Theta(r)$ has eigenvalue $1$ with right eigenvector $|\phi_1\rangle$, eigenvalues $(1-\gamma)\lambda_l$ for $l\ge 2$, right eigenvectors $|\phi_l\rangle-\frac{\gamma}{1-(1-\gamma)\lambda_l}\frac{\langle r|\phi_l\rangle}{\langle r|\phi_1\rangle}|\phi_1\rangle$, and left eigenvectors $\langle\bar\phi_l|$ except for the stationary one, which picks up a sum of corrections. The paper then writes the stationary distribution under resetting as $\frac{d_j}{N}+\gamma\sum_{l=2}^n\frac{\langle r|\phi_l\rangle\langle\bar\phi_l|j\rangle}{1-(1-\gamma)\lambda_l}$, the occupation probability as a one-line spectral sum, and the mean first passage time as Eq. (60). For the optimal reset rate, Section V derives, under the simplification $r=i$ and $i\neq j$, the condition $z^2(\gamma^*)=1+1/\langle T_{ij}(\gamma^*)\rangle$, where $z$ is the coefficient of variation of the first-passage time, together with the sufficient condition $z^2(0)>1+1/\langle T_{ij}(0)\rangle$ for resetting to be beneficial at small $\gamma$.

Load-bearing premise

The optimal-reset condition $z^2(\gamma^*)=1+1/\langle T_{ij}(\gamma^*)\rangle$ is derived only when the reset node equals the starting node and $i\neq j$, but the paper states it as a general condition for arbitrary reset nodes or reset-candidate subsets; that unproved generality is the load-bearing premise.

Editorial extensions

If this is right

  • For any fixed hypergraph, one diagonalization of the reset-free walk $W$ is enough to evaluate every reset statistic for all $\gamma$; no matrix inverse or new spectral calculation is needed per reset rate.
  • The optimal reset rate is determined by the first-passage distribution: the best $\gamma^*$ is exactly where the relative fluctuation satisfies $z^2=1+1/\langle T\rangle$.
  • Resetting is guaranteed to improve the mean first passage time whenever the original process has enough fluctuation, $z^2(0)>1+1/\langle T(0)\rangle$, which is the signature of heavy-tailed searches.
  • Node importance in a collaboration hypergraph is not the same as in its clique expansion: the hypergraph-preserving stationary distribution rewards authors with fewer, larger collaborations relative to authors with many small collaborations.
  • On the tested 40-hyperedge network, small reset rates create a dip in cover time below the no-reset value, so resetting can be tuned to speed up full coverage even though large reset rates worsen it.

Reading between the lines

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

  • The coefficient-of-variation optimality condition likely applies beyond the proven $r=i$ case, because the proof uses only the first two moments of the first-passage distribution; a brute-force sweep on small hypergraphs with $r\neq i$ would test this before a general proof is attempted.
  • Since $\Pi$'s non-unit eigenvalues are $(1-\gamma)\lambda_l$, resetting strictly increases the spectral gap, so the reset walk is expected to mix faster in the spectral-gap sense; the paper does not discuss this corollary.
  • The exact mean-first-passage formula allows cover time to be optimized as a one-dimensional search over $\gamma$ of the maximum target MFPT, replacing the paper's simulation-based sweep with a deterministic computation.
  • The weighting $K_{ij}=\sum_\alpha(C_{\alpha\alpha}-1)e_{i\alpha}e_{j\alpha}$ is a specific choice; substituting any edge-dependent vertex weighting gives a different $W$ but the same spectral machinery, so the framework can compare higher-order transition rules without changing the resetting analysis.
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 / 5 minor

Summary. The paper studies random walks with resetting on hypergraphs using the spectral decomposition of the transition matrix. It derives exact formulas for the occupation probability, stationary distribution, and mean first passage time (Eqs. (54), (55), and (60)), establishes a spectral relation between the reset and reset-free transition matrices (Eqs. (38)--(40)), proposes a generalized hypergraph transition matrix, and gives conditions for optimal resetting in terms of the coefficient of variation (Eqs. (83) and (87)). It also presents applications to node ranking and cover time. The core spectral algebra is self-contained and internally consistent; the main concern is that the optimal-resetting condition is derived only for the special case r = i, i ≠ j, although the abstract and introduction present it as a general condition.

Significance. If the results are correct, the paper provides a useful spectral toolkit for reset processes on hypergraphs: all key quantities are expressed directly in terms of the eigenvalues and eigenvectors of the reset-free transition matrix, with no fitted parameters. The relationship between the spectra of the reset and reset-free matrices is cleanly derived and checkable. The proposed generalized transition matrix and the node-ranking experiments illustrate a genuinely hypergraph-native alternative to clique expansion. The optimal-resetting conditions, once properly scoped to the r = i setting, connect to the known restart literature and are a useful discrete-time refinement. The paper is largely a theoretical contribution; the experimental sections are illustrative rather than exhaustive.

major comments (3)
  1. [Section V, Eqs. (61)--(83)] The optimal-resetting criterion is derived only under the restriction r = i and i ≠ j, stated at the beginning of Section V. The key simplification in Eq. (79) relies on r = i so that ⟨r|φ_l⟩/⟨r|φ_1⟩ cancels or reduces to a simple factor; for arbitrary reset node this cancellation does not occur, and for i = j the term δ_ij in Eq. (57) changes the derivation. The abstract and Contribution (3) nevertheless call Eq. (83) a 'general condition' for the optimal reset probability. This is a load-bearing overclaim because the optimal-resetting condition is one of the headline contributions. Please either prove the general case or explicitly state the theorem as applying to r = i, i ≠ j, and adjust the abstract and introduction accordingly.
  2. [Section II and Section IV] The paper twice announces a treatment of resetting to a subset of candidate nodes. Section II states that 'the associated transition matrix is obtained in a similar manner, which is omitted here,' and Section IV says 'As mentioned above, a subset of nodes may be selected as reset candidates. This case is also discussed in the following way.' However, no subset version is actually provided anywhere in the manuscript. Either include the subset-resetting formulation or remove these promises; as written, the reader is left with an unfulfilled claim.
  3. [Section V, Eqs. (77) and (79)] As printed, Eq. (77) does not follow from Eq. (73): the second term in Eq. (77) should contain an additional factor ⟨T_ij(γ)⟩. In addition, Eq. (79) writes the eigenvector product without the overline on the left eigenvector; the correct factor should be ⟨i|ψ_l(i;γ)⟩⟨ψ̄_l(i;γ)|j⟩, and the middle expression should involve ⟨φ̄_l|j⟩ rather than ⟨φ_l|j⟩ in both terms. These are typos in the derivation of the headline condition, but they make the displayed algebraic chain incorrect as written and must be fixed before the result can be verified.
minor comments (5)
  1. [Section V, Eqs. (61)--(63)] The notation 'N∑_{l=2}' is ambiguous: it appears to mean ∑_{l=2}^N, as in standard LaTeX rendering, but it can be read as a multiplicative factor N. Please use an explicit ∑_{l=2}^{N} or ∑_{l=2}^{n} throughout Section V to avoid confusion.
  2. [Section III-B] There is a typo in the sentence 'Revisiting the first term of the occupation probability Pij(t), whcih is...': 'whcih' should be 'which'.
  3. [Section V, text after Eq. (73)] The phrase 'revisiting Eq.(21) and q.(27)' should read 'Eq. (27)' rather than 'q. (27)'.
  4. [Section V, Eq. (83)] The relationship between Eq. (83) and Reuveni's known universal condition CV = 1 at optimal restart should be discussed explicitly. In the continuous-time or large-mean limit the extra term 1/⟨T⟩ becomes negligible, but the discrete-time correction should be acknowledged.
  5. [Section VI-A] The claim that the hypergraph-based ranking is 'more reliable' is not supported by any quantitative measure such as a rank correlation or an error bar. Consider adding a quantitative comparison or softening the wording.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the spectral derivations are self-contained and no prediction reduces to a fitted input.

full rationale

The paper's central results are derived by direct spectral algebra from the definition Π(r,γ) = (1−γ)W + γΘ(r), with no fitted parameters and no target quantity used as its own input. The eigenvalues and eigenvectors of the reset transition matrix are proven in Theorems 4–7, and the occupation probability, stationary distribution, and mean first passage time in Eqs. (54), (55), and (60) follow from the spectral decomposition of Π; they do not presuppose the conclusions. The optimal-reset condition z²(γ*) = 1 + 1/⟨T(γ*)⟩ in Eq. (83) and the sufficient condition in Eq. (87) are derived from the first and second moments of the first-passage-time distribution of the same process (Eqs. (75)–(83)), not imported as an assumption. The paper explicitly states the restriction under which that derivation is performed: 'For simplicity, we assume r = i and i ≠ j', and it openly says the subset-of-reset-candidates treatment is 'omitted here'. Those are scope limitations on the claimed generality, not circularity: the restricted theorem is still derived, and the unrestricted claim is unsupported rather than self-referential. Self-citations in the reference list (e.g., Refs. [13], [25], [30], [37], [57]) appear in background or related-work contexts and are not load-bearing for the main derivations. The hypergraph weighting Kij is adopted from prior non-author work [23], and the resetting optimality criterion is independently derived rather than justified by citing Reuveni [54]. No step was found in which a prediction is equivalent by construction to an input, a fitted parameter is renamed as a prediction, or a self-citation is used to force the central claim.

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

No constants are fitted to data; gamma is a control variable. The model's degrees of freedom are a transition-weighting choice from Carletti et al. (2020) and the fixed-reset protocol. The optimality criterion is derived from the same spectral representation, so it is not an independent fit.

assumptions (5)
  • standard math Transition matrix W = D^{-1}K is diagonalizable with biorthogonal left and right eigenvectors.
    Used throughout Section III; follows because K is symmetric and W is similar to D^{-1/2}KD^{-1/2} for a connected hypergraph.
  • domain assumption The hypergraph is connected so that the eigenvalue 1 of W is unique.
    Invoked in Section III-A to set lambda_1 = 1 and to identify the stationary distribution; disconnected hypergraphs are excluded.
  • domain assumption The appropriate weighted adjacency is K_ij = sum_alpha (C_alpha,alpha - 1) e_i,alpha e_j,alpha, so transition probabilities scale with hyperedge size minus one.
    This is the model from Carletti et al. (2020) and underlies every formula; if a different weighting is intended, the derived stationary distribution and ranking results change.
  • domain assumption Resetting is to a fixed single node r with constant probability gamma, not to a subset or with time-dependent reset.
    Defined in Section II-B; the paper mentions subsets in passing but does not derive results for them.
  • ad hoc to paper The optimal reset probability is analyzed only for reset node equal to the starting node, r = i, with i != j.
    Section V opening; the paper still calls the resulting condition general.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Random walks with resetting on hypergraph." pith.science (2026). https://pith.science/paper/S2KEK3ES

@misc{pith2026250504215,
  author       = {Pith},
  title        = {Pith review of: Random walks with resetting on hypergraph},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/S2KEK3ES}},
  note         = {Machine review of arXiv:2505.04215}
}
read the original abstract

Hypergraph has been selected as a powerful candidate for characterizing higher-order networks and has received increasing attention in recent years. In this article, we study random walks with resetting on hypergraph by utilizing spectral theory. Specifically, we derive exact expressions for some fundamental yet key parameters, including occupation probability, stationary distribution, and mean first passage time, all of which are expressed in terms of the eigenvalues and eigenvectors of the transition matrix. Furthermore, we provide a general condition for determining the optimal reset probability and a sufficient condition for its existence. In addition, we build up a close relationship between random walks with resetting on hypergraph and simple random walks. Concretely, the eigenvalues and eigenvectors of the former can be precisely represented by those of the latter. More importantly, when considering random walks, we abandon the traditional approach of converting hypergraph into a graph and propose a research framework that preserves the intrinsic structure of hypergraph itself, which is based on assigning proper weights to neighboring nodes. Through extensive experiments, we show that the new framework produces distinct and more reliable results than the traditional approach in node ranking. Finally, we explore the impact of the resetting mechanism on cover time, providing a potential solution for optimizing search efficiency.

Figures

Figures reproduced from arXiv: 2505.04215 by the authors.

Figure 1
Figure 1. An example hypergraph indicating a co-authorship network where each node represents an author, and each hyperedge connects all [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The scatter plot of the normalized ranking obtained from random walks on the hypergraph. [PITH_FULL_IMAGE:figures/full_fig_p017_2.png] view at source ↗
Figure 3
Figure 3. The scatter plot of the normalized ranking obtained from random walks on the corresponding clique graph. [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Comparison of the normalized ranking P∞ i / maxj P∞ j on the hypergraph and Q∞ i / maxj Q∞ j on the corresponding clique graph. hypergraph, each hyperedge is converted into a central node, and all the nodes within the hyperedge are connected to this central node. This …
Figure 5
Figure 5. Figure 5: Cover time under different reset rates. The reset rate ranges from 0 to 0 [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

65 extracted references · 63 canonical work pages

  1. [1]

    Modeling echo chambers and polarization dynamics in social networks,

    F. Baumann, P. Lorenz-Spreen, I. M. Sokolov, and M. Starnini, “Modeling echo chambers and polarization dynamics in social networks,” Physical Review Letters, vol. 124, no. 4, p. 048301, 2020

  2. [2]

    Transfer learning enables predictions in network biology,

    C. V. Theodoris, L. Xiao, A. Chopra, M. D. Chaffin, Z. R. Al Sayed, M. C. Hill, H. Mantineo, E. M. Brydon, Z. Zeng, X. S. Liu et al., “Transfer learning enables predictions in network biology,” Nature, vol. 618, no. 7965, pp. 616–624, 2023

  3. [3]

    Data-driven analysis and forecasting of highway traffic dynamics,

    A. M. Avila and I. Mezi´ c, “Data-driven analysis and forecasting of highway traffic dynamics,” Nature Communications, vol. 11, no. 1, p. 2090, 2020

  4. [4]

    Understanding influence of fractal generative manner on structural properties of tree networks,

    F. Ma and P. Wang, “Understanding influence of fractal generative manner on structural properties of tree networks,” Chaos, Solitons & Fractals, vol. 180, p. 114523, 2024

  5. [5]

    What are higher-order networks?

    C. Bick, E. Gross, H. A. Harrington, and M. T. Schaub, “What are higher-order networks?” SIAM Review, vol. 65, no. 3, pp. 686–731, 2023

  6. [6]

    The why, how, and when of representations for complex systems,

    L. Torres, A. S. Blevins, D. Bassett, and T. Eliassi-Rad, “The why, how, and when of representations for complex systems,” SIAM Review, vol. 63, no. 3, pp. 435–485, 2021

  7. [7]

    Networks beyond pairwise interactions: Structure and dynamics,

    F. Battiston, G. Cencetti, I. Iacopini, V. Latora, M. Lucas, A. Patania, J.-G. Young, and G. Petri, “Networks beyond pairwise interactions: Structure and dynamics,” Physics Reports, vol. 874, pp. 1–92, 2020

  8. [8]

    Nonuniform hyper-network embedding with dual mechanism,

    J. Huang, C. Chen, F. Ye, W. Hu, and Z. Zheng, “Nonuniform hyper-network embedding with dual mechanism,” ACM Transactions on Information Systems (TOIS) , vol. 38, no. 3, pp. 1–18, 2020

Show all 65 references
  1. [9]

    Species-specific segmentation clock periods are due to differential biochemical reaction speeds,

    M. Matsuda, H. Hayashi, J. Garcia-Ojalvo, K. Yoshioka-Kobayashi, R. Kageyama, Y. Yamanaka, M. Ikeya, J. Toguchida, C. Alev, and M. Ebisuya, “Species-specific segmentation clock periods are due to differential biochemical reaction speeds,” Science, vol. 369, no. 6510, pp. 1450–...

  2. [10]

    A survey on hypergraph representation learning,

    A. Antelmi, G. Cordasco, M. Polato, V. Scarano, C. Spagnuolo, and D. Yang, “A survey on hypergraph representation learning,” ACM Computing Surveys, vol. 56, no. 1, pp. 1–38, 2023. JOURNAL OF LATEX CLASS FILES, VOL. XX, NO. XX, XX 202X 21

  3. [11]

    Statistical physics of social dynamics,

    C. Castellano, S. Fortunato, and V. Loreto, “Statistical physics of social dynamics,” Reviews of Modern Physics , vol. 81, no. 2, pp. 591–646, 2009

  4. [12]

    Synchronization in complex networks,

    A. Arenas, A. D´ ıaz-Guilera, J. Kurths, Y. Moreno, and C. Zhou, “Synchronization in complex networks,” Physics Reports, vol. 469, no. 3, pp. 93–153, 2008

  5. [13]

    Random walks on a tree with applications,

    F. Ma and P. Wang, “Random walks on a tree with applications,” Physical Review E, vol. 102, no. 2, p. 022305, 2020

  6. [14]

    G. F. Lawler and V. Limic, Random walk: a modern introduction. Cambridge University Press, 2010

  7. [15]

    Structural properties on scale-free tree network with an ultra-large diameter,

    F. Ma and P. Wang, “Structural properties on scale-free tree network with an ultra-large diameter,” ACM Transactions on Knowledge Discovery from Data, vol. 18, no. 8, pp. 1–26, 2024

  8. [16]

    Random walks on complex networks,

    J. D. Noh and H. Rieger, “Random walks on complex networks,” Physical Review Letters, vol. 92, no. 11, p. 118701, 2004

  9. [17]

    Random walks on graphs,

    L. Lov´ asz, “Random walks on graphs,”Combinatorics, Paul Erdos is Eighty , vol. 2, no. 1-46, p. 4, 1993

  10. [18]

    A measure of betweenness centrality based on random walks,

    M. E. Newman, “A measure of betweenness centrality based on random walks,” Social Networks, vol. 27, no. 1, pp. 39–54, 2005

  11. [19]

    Maps of random walks on complex networks reveal community structure,

    M. Rosvall and C. T. Bergstrom, “Maps of random walks on complex networks reveal community structure,”Proceedings of the National Academy of Sciences, vol. 105, no. 4, pp. 1118–1123, 2008

  12. [20]

    Characteristic exponents of complex networks,

    V. Nicosia, M. De Domenico, and V. Latora, “Characteristic exponents of complex networks,” Europhysics Letters, vol. 106, no. 5, p. 58005, 2014

  13. [21]

    Hypergraph random walks, laplacians, and clustering,

    K. Hayashi, S. G. Aksoy, C. H. Park, and H. Park, “Hypergraph random walks, laplacians, and clustering,” in Proceedings of the 29th ACM International Conference on Information & Knowledge Management , 2020, pp. 495–504

  14. [22]

    Random walks on hypergraphs with edge-dependent vertex weights,

    U. Chitra and B. Raphael, “Random walks on hypergraphs with edge-dependent vertex weights,” in International Conference on Machine Learning. PMLR, 2019, pp. 1172–1181

  15. [23]

    Random walks on hypergraphs,

    T. Carletti, F. Battiston, G. Cencetti, and D. Fanelli, “Random walks on hypergraphs,” Physical Review E, vol. 101, no. 2, p. 022308, 2020

  16. [24]

    Dynamics of nonlinear random walks on complex networks,

    P. S. Skardal and S. Adhikari, “Dynamics of nonlinear random walks on complex networks,” Journal of Nonlinear Science, vol. 29, pp. 1419–1444, 2019

  17. [25]

    Structure diversity and mean hitting time for random walks on stochastic uniform growth tree networks,

    F. Ma, P. Wang, X. Luo, and R. Zhu, “Structure diversity and mean hitting time for random walks on stochastic uniform growth tree networks,” IEEE Transactions on Knowledge and Data Engineering , vol. 35, no. 8, pp. 8572–8583, 2023

  18. [26]

    Reactive random walkers on complex networks,

    G. Cencetti, F. Battiston, D. Fanelli, and V. Latora, “Reactive random walkers on complex networks,” Physical Review E, vol. 98, no. 5, p. 052302, 2018

  19. [27]

    Random walks on networks with stochastic resetting,

    A. P. Riascos, D. Boyer, P. Herringer, and J. L. Mateos, “Random walks on networks with stochastic resetting,” Physical Review E, vol. 101, no. 6, p. 062147, 2020

  20. [28]

    Random walk with restart: fast solutions and applications,

    H. Tong, C. Faloutsos, and J.-Y. Pan, “Random walk with restart: fast solutions and applications,”Knowledge and Information Systems, vol. 14, pp. 327–346, 2008

  21. [29]

    Diffusion with stochastic resetting,

    M. R. Evans and S. N. Majumdar, “Diffusion with stochastic resetting,” Physical Review Letters, vol. 106, no. 16, p. 160601, 2011

  22. [30]

    Random walks on stochastic generalized vicsek fractal networks: Analytic solution and simulations,

    F. Ma, P. Wang, and X. Luo, “Random walks on stochastic generalized vicsek fractal networks: Analytic solution and simulations,” IEEE Transactions on Network Science and Engineering , vol. 9, no. 3, pp. 1335–1345, 2022

  23. [31]

    Optimal speedup of las vegas algorithms,

    M. Luby, A. Sinclair, and D. Zuckerman, “Optimal speedup of las vegas algorithms,” Information Processing Letters, vol. 47, no. 4, pp. 173–180, 1993

  24. [32]

    Optimizing searches via rare events,

    A. Montanari and R. Zecchina, “Optimizing searches via rare events,” Physical Review Letters, vol. 88, no. 17, p. 178701, 2002

  25. [33]

    Random walks with preferential relocations to places visited in the past and their application to biology,

    D. Boyer and C. Solis-Salas, “Random walks with preferential relocations to places visited in the past and their application to biology,” Physical Review Letters, vol. 112, no. 24, p. 240601, 2014

  26. [34]

    Simrank: a measure of structural-context similarity,

    G. Jeh and J. Widom, “Simrank: a measure of structural-context similarity,” in Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 2002, pp. 538–543

  27. [35]

    Network-based recommendation algorithms: A review,

    F. Yu, A. Zeng, S. Gillard, and M. Medo, “Network-based recommendation algorithms: A review,” Physica A: Statistical Mechanics and Its Applications, vol. 452, pp. 192–208, 2016

  28. [36]

    Dblp: some lessons learned,

    M. Ley, “Dblp: some lessons learned,” Proceedings of the VLDB Endowment, vol. 2, no. 2, pp. 1493–1500, 2009

  29. [37]

    Random walks on the generalized vicsek fractal,

    F. Ma, X. Wang, P. Wang, and X. Luo, “Random walks on the generalized vicsek fractal,” Europhysics Letters, vol. 133, no. 4, p. 40004, 2021

  30. [38]

    Random walks on fibonacci treelike models,

    F. Ma, P. Wang, and B. Yao, “Random walks on fibonacci treelike models,” Physica A: Statistical Mechanics and its Applications, vol. 581, p. 126199, 2021

  31. [39]

    D. J. Griffiths and D. F. Schroeter, Introduction to quantum mechanics. Cambridge University Press, 2018

  32. [40]

    Community structure in social and biological networks,

    M. Girvan and M. E. Newman, “Community structure in social and biological networks,” Proceedings of the National Academy of Sciences, vol. 99, no. 12, pp. 7821–7826, 2002

  33. [41]

    Centrality and network flow,

    S. P. Borgatti, “Centrality and network flow,” Social Networks, vol. 27, no. 1, pp. 55–71, 2005

  34. [42]

    Identifying influential and susceptible members of social networks,

    S. Aral and D. Walker, “Identifying influential and susceptible members of social networks,” Science, vol. 337, no. 6092, pp. 337–341, 2012

  35. [43]

    Maximizing the spread of influence through a social network,

    D. Kempe, J. Kleinberg, and ´E. Tardos, “Maximizing the spread of influence through a social network,” in Proceedings of the Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 2003, pp. 137–146

  36. [44]

    Identification of influential spreaders in complex networks,

    M. Kitsak, L. K. Gallos, S. Havlin, F. Liljeros, L. Muchnik, H. E. Stanley, and H. A. Makse, “Identification of influential spreaders in complex networks,” Nature Physics, vol. 6, no. 11, pp. 888–893, 2010

  37. [45]

    Catastrophic cascade of failures in interdependent networks,

    S. V. Buldyrev, R. Parshani, G. Paul, H. E. Stanley, and S. Havlin, “Catastrophic cascade of failures in interdependent networks,” Nature, vol. 464, no. 7291, pp. 1025–1028, 2010

  38. [46]

    The anatomy of a large-scale hypertextual web search engine,

    S. Brin and L. Page, “The anatomy of a large-scale hypertextual web search engine,” Computer Networks and ISDN Systems , vol. 30, no. 1-7, pp. 107–117, 1998

  39. [47]

    Authoritative sources in a hyperlinked environment,

    J. M. Kleinberg, “Authoritative sources in a hyperlinked environment,” Journal of the ACM (JACM), vol. 46, no. 5, pp. 604–632, 1999

  40. [48]

    Topic-sensitive pagerank,

    T. H. Haveliwala, “Topic-sensitive pagerank,” in Proceedings of the 11th International Conference on World Wide Web , 2002, pp. 517–526

  41. [49]

    Graph convolutional neural networks for web-scale recommender systems,

    R. Ying, R. He, K. Chen, P. Eksombatchai, W. L. Hamilton, and J. Leskovec, “Graph convolutional neural networks for web-scale recommender systems,” in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data mining , 2018, pp. 974–983

  42. [50]

    Impact of human mobility on opportunistic forwarding algorithms,

    A. Chaintreau, P. Hui, J. Crowcroft, C. Diot, R. Gass, and J. Scott, “Impact of human mobility on opportunistic forwarding algorithms,” IEEE Transactions on Mobile Computing , vol. 6, no. 6, pp. 606–620, 2007

  43. [51]

    Secure data collection in wireless sensor networks using randomized dispersive routes,

    T. Shu, M. Krunz, and S. Liu, “Secure data collection in wireless sensor networks using randomized dispersive routes,” IEEE Transactions on Mobile Computing, vol. 9, no. 7, pp. 941–954, 2010

  44. [52]

    Cover times for brownian motion and random walks in two dimensions,

    A. Dembo, Y. Peres, J. Rosen, and O. Zeitouni, “Cover times for brownian motion and random walks in two dimensions,” Annals of Mathematics, pp. 433–464, 2004

  45. [53]

    Diffusion with optimal resetting,

    M. R. Evans and S. N. Majumdar, “Diffusion with optimal resetting,” Journal of Physics A: Mathematical and Theoretical , vol. 44, no. 43, p. 435001, 2011. JOURNAL OF LATEX CLASS FILES, VOL. XX, NO. XX, XX 202X 22

  46. [54]

    Optimal stochastic restart renders fluctuations in first passage times universal,

    S. Reuveni, “Optimal stochastic restart renders fluctuations in first passage times universal,” Physical Review Letters, vol. 116, no. 17, p. 170601, 2016

  47. [55]

    Stochastic resetting and applications,

    M. R. Evans, S. N. Majumdar, and G. Schehr, “Stochastic resetting and applications,” Journal of Physics A: Mathematical and Theoretical, vol. 53, no. 19, p. 193001, 2020

  48. [56]

    The structure and function of complex networks,

    M. E. Newman, “The structure and function of complex networks,” SIAM Review, vol. 45, no. 2, pp. 167–256, 2003

  49. [57]

    Determining mean first-passage time for random walks on stochastic uniform growth tree networks,

    F. Ma and P. Wang, “Determining mean first-passage time for random walks on stochastic uniform growth tree networks,” IEEE Transactions on Knowledge and Data Engineering , 2024

  50. [58]

    The cover time, the blanket time, and the matthews bound,

    J. Kahn, J. H. Kim, L. Lovasz, and V. H. Vu, “The cover time, the blanket time, and the matthews bound,” in Proceedings 41st Annual Symposium on Foundations of Computer Science . IEEE, 2000, pp. 467–475

  51. [59]

    Learning with hypergraphs: Clustering, classification, and embedding,

    D. Zhou, J. Huang, and B. Sch¨ olkopf, “Learning with hypergraphs: Clustering, classification, and embedding,” Advances in Neural Information Processing Systems, vol. 19, 2006

  52. [60]

    Higher order learning with graphs,

    S. Agarwal, K. Branson, and S. Belongie, “Higher order learning with graphs,” in Proceedings of the 23rd International Conference on Machine Learning, 2006, pp. 17–24

  53. [61]

    An inverse power method for nonlinear eigenproblems with applications in 1-spectral clustering and sparse pca,

    M. Hein and T. B¨ uhler, “An inverse power method for nonlinear eigenproblems with applications in 1-spectral clustering and sparse pca,” Advances in Neural Information Processing Systems, vol. 23, 2010

  54. [62]

    Beyond pairwise clustering,

    S. Agarwal, J. Lim, L. Zelnik-Manor, P. Perona, D. Kriegman, and S. Belongie, “Beyond pairwise clustering,” in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05) , vol. 2. IEEE, 2005, pp. 838–845

  55. [63]

    A fast and high quality multilevel scheme for partitioning irregular graphs,

    G. Karypis and V. Kumar, “A fast and high quality multilevel scheme for partitioning irregular graphs,” SIAM Journal on Scientific Computing, vol. 20, no. 1, pp. 359–392, 1998

  56. [64]

    Semi-supervised hypergraph node classification on hypergraph line expansion,

    C. Yang, R. Wang, S. Yao, and T. Abdelzaher, “Semi-supervised hypergraph node classification on hypergraph line expansion,” in Proceedings of the 31st ACM International Conference on Information & Knowledge Management , 2022, pp. 2352–2361

  57. [65]

    Hypergraph attention isomorphism network by learning line graph expansion,

    S. Bandyopadhyay, K. Das, and M. N. Murty, “Hypergraph attention isomorphism network by learning line graph expansion,” in 2020 IEEE International Conference on Big Data (Big Data) . IEEE, 2020, pp. 669–678

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.