pith. sign in

arxiv: 2605.28300 · v1 · pith:S7QF2GOEnew · submitted 2026-05-27 · 💻 cs.LG

T-GINEE: A Tensor-Based Multilayer Graph Representation Learning

Maolin Wang , Ziting Mai , Xuhui Chen , Zhiqi Li , Tianshuo Wei , Yutian Xiao , Wenlin Zhang , Wanyu Wang
show 4 more authors
Ruocheng Guo Haoxuan Li Zenglin Xu Xiangyu Zhao
This is my paper

Pith reviewed 2026-06-29 14:12 UTC · model grok-4.3

classification 💻 cs.LG
keywords multilayer networkstensor decompositiongeneralized estimating equationsgraph representation learningcross-layer correlationsstatistical regularizationconsistency analysis
0
0 comments X

The pith

T-GINEE models cross-layer correlations in multilayer networks through CP tensor decomposition inside a generalized estimating equation framework.

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

The paper develops T-GINEE to analyze networks that contain multiple layers, each representing a different type of relationship. Standard methods either study each layer on its own or merge layers without regard to how they influence one another. T-GINEE instead decomposes the multilayer structure into shared latent factors and uses estimating equations to encode the statistical dependence between layers. This produces node representations that respect the full joint structure rather than isolated slices. Readers would care because transportation systems, social platforms, and biological interaction data routinely appear as such interdependent layers.

Core claim

T-GINEE is a statistical regularization framework that combines CP tensor decomposition to capture structural dependencies via shared latent factors, a generalized estimating equation framework that models inter-layer correlations through working covariance matrices, and a flexible link function that accommodates traits such as sparsity. The method adds a task-specific loss and supplies theoretical results establishing consistency and asymptotic normality of the estimators under mild conditions. Experiments on both synthetic and real multilayer datasets show improved performance relative to baselines that ignore cross-layer dependence.

What carries the argument

CP tensor decomposition paired with the generalized estimating equation framework whose working covariance matrices encode inter-layer dependence.

If this is right

  • Node embeddings respect explicit cross-layer statistical dependence rather than assuming layer independence.
  • The estimators remain consistent and asymptotically normal whenever the mild regularity conditions are satisfied.
  • The same framework can be paired with different task losses for link prediction, node classification, or community detection on multilayer data.
  • Sparsity and other layer-specific traits are handled directly by the choice of link function.

Where Pith is reading between the lines

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

  • If the working covariance matrices are badly misspecified, the method may lose its statistical guarantees, pointing to a need for data-driven covariance estimation.
  • The tensor-plus-GEE structure could be extended to time-varying multilayer networks by adding a temporal smoothness term.
  • Similar tensor decompositions appear in other network models; comparing the resulting estimators might reveal when the GEE layer adds value beyond pure tensor factorization.

Load-bearing premise

The working covariance matrices chosen inside the generalized estimating equation framework are close enough to the true inter-layer correlation structure.

What would settle it

A controlled simulation in which the true inter-layer correlation matrix is known and deliberately mismatched to the working covariance matrices, then checking whether the claimed consistency and asymptotic normality still hold.

Figures

Figures reproduced from arXiv: 2605.28300 by Haoxuan Li, Maolin Wang, Ruocheng Guo, Tianshuo Wei, Wanyu Wang, Wenlin Zhang, Xiangyu Zhao, Xuhui Chen, Yutian Xiao, Zenglin Xu, Zhiqi Li, Ziting Mai.

Figure 1
Figure 1. Figure 1: Schematic overview of the T-GINEE framework. Given a multilayer adjacency tensor A, we introduce a parameter tensor Θ and perform a symmetric CP decomposition to obtain node embeddings α and layer-specific embeddings β. These embeddings are concatenated into a parameter vector γ. By solving tensor-based generalized estimating equations (T-GINEE) under a working covariance structure, the model learns γ and … view at source ↗
Figure 2
Figure 2. Figure 2: Sensitivity of T-GINEE to graph sparsity on synthetic multilayer networks. We vary the proportion of observed edges (proportion of 1 entries) and report the mean test AUC over 5 runs with one standard deviation as the shaded region. walk-based approaches, including DeepWalk (Perozzi et al., 2014) and node2vec (Grover & Leskovec, 2016), adapt word embedding techniques to networks. More recently, graph convo… view at source ↗
Figure 3
Figure 3. Figure 3: Comprehensive analysis of model hyperparameters: (a) embedding dimension impact, (b) regularization effect, and (c) computational efficiency. Impact of embedding dimension. As shown in Figure 3a, the relationship between embedding dimension and model performance demonstrates clear patterns across different regularization settings. The experimental results show that increasing the embedding dimension genera… view at source ↗
read the original abstract

Traditional network analysis focuses on single-layer networks, real-world systems often form multilayer networks with multiple relationship types. However, existing methods typically fail to capture complex inter-layer dependencies by treating layers independently or aggregating them. To address this, we propose T-GINEE (Tensor-Based Generalized Multilayer-graph Estimating Equation), a statistical regularization framework combining tensor-based generalized estimating equations with task-specific loss to model cross-network correlations explicitly. Key innovations include: (1) CP tensor decomposition capturing structural dependencies via shared latent factors; (2) a generalized estimating equation framework modeling inter-layer correlations through working covariance matrices; and (3) a flexible link function accommodating characteristics like sparsity. Our theoretical analysis establishes consistency and asymptotic normality under mild conditions. Extensive experiments on synthetic and real-world datasets validate T-GINEE's effectiveness for multilayer network analysis.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 1 minor

Summary. The paper proposes T-GINEE, a statistical regularization framework that integrates CP tensor decomposition with generalized estimating equations (GEE) and a task-specific loss to explicitly model inter-layer correlations in multilayer networks. It introduces shared latent factors via CP decomposition, working covariance matrices for correlations, and a flexible link function for sparsity, while claiming consistency and asymptotic normality of estimators under mild conditions, validated through experiments on synthetic and real-world datasets.

Significance. If the theoretical results hold with verifiable conditions, the work offers a statistically grounded alternative to independent or aggregated layer treatments in multilayer graph learning by directly incorporating cross-network dependencies through tensor-GEE machinery. The explicit use of GEE working covariances and CP factors for structural dependencies is a potentially useful synthesis, though its advantage depends on whether the claimed guarantees transfer to typical graph sparsity regimes.

major comments (1)
  1. [Theoretical Analysis] Theoretical Analysis (or equivalent section containing the consistency/asymptotic normality proofs): The central claim of consistency and asymptotic normality under 'mild conditions' is load-bearing for the paper's contribution, yet the abstract and framework description provide no explicit statement of these conditions (e.g., requirements on moments, identifiability of CP rank, correctness of the link function, or boundedness of the working covariance matrices). Standard GEE theory requires correct mean specification for consistency but ties asymptotic normality and efficiency to the working covariance accurately reflecting inter-layer dependence; without verifying these for multilayer graph sparsity patterns or finite-sample regimes, the guarantees do not demonstrably apply.
minor comments (1)
  1. [Abstract] Abstract: The description of 'task-specific loss' and 'flexible link function' is high-level; a concrete example of the link function (e.g., logit or identity) and how it interacts with the CP factors would improve clarity.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed and constructive feedback on our manuscript. The concern regarding explicit conditions in the theoretical analysis is well-taken, and we address it directly below with a commitment to revision.

read point-by-point responses

  1. Referee: Theoretical Analysis (or equivalent section containing the consistency/asymptotic normality proofs): The central claim of consistency and asymptotic normality under 'mild conditions' is load-bearing for the paper's contribution, yet the abstract and framework description provide no explicit statement of these conditions (e.g., requirements on moments, identifiability of CP rank, correctness of the link function, or boundedness of the working covariance matrices). Standard GEE theory requires correct mean specification for consistency but ties asymptotic normality and efficiency to the working covariance accurately reflecting inter-layer dependence; without verifying these for multilayer graph sparsity patterns or finite-sample regimes, the guarantees do not demonstrably apply.

    Authors: We agree that the abstract and framework overview do not enumerate the conditions in detail, which reduces transparency. The proofs in the theoretical analysis section rely on standard GEE regularity conditions adapted to the CP tensor decomposition: (i) finite second moments of the multilayer observations, (ii) identifiability of the CP rank under the assumed decomposition, (iii) correct specification of the conditional mean via the chosen link function, and (iv) bounded eigenvalues of the working covariance matrices ensuring they remain positive definite. To make these explicit, we will add a dedicated 'Assumptions' subsection immediately preceding the consistency and asymptotic normality theorems, listing each condition with precise mathematical statements and references to the relevant lemmas. We will also expand the discussion to address multilayer graph sparsity by noting that the link function (e.g., logit) and the working covariance construction are chosen to be compatible with sparse regimes where edge probabilities decay appropriately, preserving the mean-correctness requirement. For finite-sample behavior, the results are asymptotic; we will add a remark clarifying this and referencing the synthetic experiments that empirically support the theory in moderate-sized sparse graphs. These changes will be incorporated in the revised manuscript. revision: yes

Circularity Check

0 steps flagged

No significant circularity; framework integrates established tensor and GEE methods without self-referential reductions

full rationale

The described T-GINEE framework combines CP tensor decomposition to capture structural dependencies via shared latent factors, a generalized estimating equation setup with working covariance matrices for inter-layer correlations, and a flexible link function. Theoretical consistency and asymptotic normality are asserted under mild conditions, but the provided abstract and description contain no equations or steps where a prediction or result reduces by construction to a fitted parameter, self-defined quantity, or self-citation chain. The derivation remains self-contained as an application of standard statistical techniques to multilayer graphs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; the method appears to rest on standard assumptions from tensor decomposition and generalized estimating equations plus unspecified mild conditions for the asymptotic results.

axioms (1)
  • domain assumption Mild conditions suffice for consistency and asymptotic normality of the estimators
    Invoked for the theoretical analysis described in the abstract.

pith-pipeline@v0.9.1-grok · 5708 in / 1167 out tokens · 28516 ms · 2026-06-29T14:12:07.854645+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

66 extracted references · 11 canonical work pages · 3 internal anchors

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

  2. [2]

    A tensor factorization model of multilayer network interdependence, 2024

    Aguiar, I., Taylor, D., and Ugander, J. A tensor factorization model of multilayer network interdependence, 2024. URL https://arxiv.org/abs/2206.01804

    work page arXiv 2024

  3. [3]

    G., Mangioni, G., Cingolani, I., Rodrigues, F

    Alves, L. G., Mangioni, G., Cingolani, I., Rodrigues, F. A., Panzarasa, P., and Moreno, Y. The nested structural organization of the worldwide trade multi-layer network. Scientific reports, 9 0 (1): 0 2866, 2019

    2019

  4. [4]

    Ash, R. B. and Dol \'e ans-Dade, C. A. Probability and measure theory. Academic press, 2000

    2000

  5. [5]

    Group formation in large social networks: membership, growth, and evolution

    Backstrom, L., Huttenlocher, D., Kleinberg, J., and Lan, X. Group formation in large social networks: membership, growth, and evolution. In Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, pp.\ 44--54, 2006

    2006

  6. [6]

    I., G \'o mez-Gardenes, J., Romance, M., Sendiña-Nadal, I., Wang, Z., and Zanin, M

    Boccaletti, S., Bianconi, G., Criado, R., Del Genio, C. I., G \'o mez-Gardenes, J., Romance, M., Sendiña-Nadal, I., Wang, Z., and Zanin, M. The structure and dynamics of multilayer networks. Physics Reports, 544 0 (1): 0 1--122, 2014

    2014

  7. [7]

    Mimag: mining coherent subgraphs in multi-layer graphs with edge labels

    Boden, B., G \"u nnemann, S., Hoffmann, H., and Seidl, T. Mimag: mining coherent subgraphs in multi-layer graphs with edge labels. Knowledge and Information Systems, 50 0 (2): 0 417--446, 2017

    2017

  8. [8]

    Brown, B. M. Martingale central limit theorems. The Annals of Mathematical Statistics, pp.\ 59--66, 1971

    1971

  9. [9]

    Cai, D., He, X., Han, J., and Huang, T. S. Graph regularized nonnegative matrix factorization for data representation. IEEE transactions on pattern analysis and machine intelligence, 33 0 (8): 0 1548--1560, 2010

    2010

  10. [10]

    W., and Chang, K

    Cai, H., Zheng, V. W., and Chang, K. C.-C. A comprehensive survey of graph embedding: Problems, techniques, and applications. IEEE transactions on knowledge and data engineering, 30 0 (9): 0 1616--1637, 2018

    2018

  11. [11]

    A., G \'o mez, S., and Arenas, A

    De Domenico, M., Sol \'e -Ribalta, A., Cozzo, E., Kivel \"a , M., Moreno, Y., Porter, M. A., G \'o mez, S., and Arenas, A. Mathematical formulation of multilayer networks. Physical Review X, 3 0 (4): 0 041022, 2013

    2013

  12. [12]

    Structural reducibility of multilayer networks

    De Domenico, M., Nicosia, V., Arenas, A., and Latora, V. Structural reducibility of multilayer networks. Nature Communications, 6: 0 6864, 04 2015. doi:10.1038/ncomms7864

    work page doi:10.1038/ncomms7864 2015

  13. [13]

    Clustering with multi-layer graphs: A spectral perspective

    Dong, X., Frossard, P., Vandergheynst, P., and Nefedov, N. Clustering with multi-layer graphs: A spectral perspective. IEEE Transactions on Signal Processing, 60 0 (11): 0 5820--5831, 2012

    2012

  14. [14]

    V., and Swami, A

    Dong, Y., Chawla, N. V., and Swami, A. metapath2vec: Scalable representation learning for heterogeneous networks. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.\ 135--144, 2017

    2017

  15. [15]

    S., and Rabiee, H

    Ghorbani, M., Baghshah, M. S., and Rabiee, H. R. Mgcn: semi-supervised classification in multi-layer graphs with graph convolutional networks. In Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, pp.\ 208--211, 2019

    2019

  16. [16]

    Fusion and community detection in multi-layer graphs

    Gligorijevi \'c , V., Panagakis, Y., and Zafeiriou, S. Fusion and community detection in multi-layer graphs. In 2016 23rd international conference on pattern recognition (ICPR), pp.\ 1327--1332. IEEE, 2016

    2016

  17. [17]

    Golub, G. H. and Reinsch, C. Singular value decomposition and least squares solutions. In Linear algebra, pp.\ 134--151. Springer, 1971

    1971

  18. [18]

    and Cunningham, P

    Greene, D. and Cunningham, P. Producing a unified graph representation from multiple social network views. In Proceedings of the 5th annual ACM web science conference, pp.\ 118--121, 2013

    2013

  19. [19]

    and Leskovec, J

    Grover, A. and Leskovec, J. node2vec: Scalable feature learning for networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.\ 855--864, 2016

    2016

  20. [20]

    Inductive representation learning on large graphs

    Hamilton, W., Ying, Z., and Leskovec, J. Inductive representation learning on large graphs. Advances in neural information processing systems, 30, 2017

    2017

  21. [21]

    Consistent estimation of dynamic and multi-layer block models

    Han, Q., Xu, K., and Airoldi, E. Consistent estimation of dynamic and multi-layer block models. In International Conference on Machine Learning, pp.\ 1511--1520. PMLR, 2015

    2015

  22. [22]

    Huang, Z., Li, X., Ye, Y., and Ng, M. K. Mr-gcn: Multi-relational graph convolutional networks based on generalized tensor product. In IJCAI, volume 20, pp.\ 1258--1264, 2020

    2020

  23. [23]

    Multilayer network simplification: approaches, models and methods

    Interdonato, R., Magnani, M., Perna, D., Tagarelli, A., and Vega, D. Multilayer network simplification: approaches, models and methods. Computer Science Review, 36: 0 100246, 2020

    2020

  24. [24]

    An effective and robust framework by modeling correlations of multiplex network embedding

    Jiao, P., Lu, R., Jin, D., Wang, Y., and Wu, H. An effective and robust framework by modeling correlations of multiplex network embedding. In 2021 IEEE International Conference on Data Mining (ICDM), pp.\ 1144--1149. IEEE, 2021

    2021

  25. [25]

    Community detection on mixture multilayer networks via regularized tensor decomposition

    Jing, B.-Y., Li, T., Lyu, Z., and Xia, D. Community detection on mixture multilayer networks via regularized tensor decomposition. The Annals of Statistics, 49 0 (6): 0 3181--3205, 2021

    2021

  26. [26]

    Kipf, T. N. and Welling, M. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907, 2016

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  27. [27]

    P., Moreno, Y., and Porter, M

    Kivel \"a , M., Arenas, A., Barthelemy, M., Gleeson, J. P., Moreno, Y., and Porter, M. A. Multilayer networks. Journal of Complex Networks, 2 0 (3): 0 203--271, 2014

    2014

  28. [28]

    Kolda, T. G. and Bader, B. W. Tensor decompositions and applications. SIAM review, 51 0 (3): 0 455--500, 2009

    2009

  29. [29]

    Cognitive social structures

    Krackhardt, D. Cognitive social structures. Social Networks, 9 0 (2): 0 109--134, 1987. ISSN 0378-8733. doi:https://doi.org/10.1016/0378-8733(87)90009-8

    work page doi:10.1016/0378-8733(87)90009-8 1987

  30. [30]

    and Seung, H

    Lee, D. and Seung, H. S. Algorithms for non-negative matrix factorization. Advances in neural information processing systems, 13, 2000

    2000

  31. [31]

    Consistent community detection in multi-layer network data

    Lei, J., Chen, K., and Lynch, B. Consistent community detection in multi-layer network data. Biometrika, 107 0 (1): 0 61--73, 2020

    2020

  32. [32]

    Active module identification from multilayer weighted gene co-expression networks: a continuous optimization approach

    Li, D., Pan, Z., Hu, G., Anderson, G., and He, S. Active module identification from multilayer weighted gene co-expression networks: a continuous optimization approach. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 18 0 (6): 0 2239--2248, 2020

    2020

  33. [33]

    and Zeger, S

    Liang, K.-Y. and Zeger, S. L. Longitudinal data analysis using generalized linear models. biometrika, pp.\ 13--22, 1986

    1986

  34. [34]

    Principled multilayer network embedding

    Liu, W., Chen, P.-Y., Yeung, S., Suzumura, T., and Chen, L. Principled multilayer network embedding. In 2017 IEEE International Conference on Data Mining Workshops (ICDMW), pp.\ 134--141. IEEE, 2017

    2017

  35. [35]

    B., Loscalzo, J., Gao, J., and Sharma, A

    Liu, X., Maiorino, E., Halu, A., Glass, K., Prasad, R. B., Loscalzo, J., Gao, J., and Sharma, A. Robustness and lethality in multilayer biological molecular networks. Nature communications, 11 0 (1): 0 6043, 2020

    2020

  36. [36]

    Pretraining representations of multi-modal multi-query e-commerce search

    Liu, X., Guan, W., Li, L., Li, H., Lin, C., Li, X., Chen, S., Xu, J., Deng, H., and Zheng, B. Pretraining representations of multi-modal multi-query e-commerce search. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pp.\ 3429--3437, 2022

    2022

  37. [37]

    Latent space model for higher-order networks and generalized tensor decomposition

    Lyu, Z., Xia, D., and Zhang, Y. Latent space model for higher-order networks and generalized tensor decomposition. Journal of Computational and Graphical Statistics, 32 0 (4): 0 1320--1336, 2023

    2023

  38. [38]

    and Tapper, U

    Paatero, P. and Tapper, U. Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values. Environmetrics, 5 0 (2): 0 111--126, 1994

    1994

  39. [39]

    E., Akoglu, L., and Ience, D

    Papalexakis, E. E., Akoglu, L., and Ience, D. Do more views of a graph help? community detection and clustering in multi-graphs. In Proceedings of the 16th International Conference on Information Fusion, pp.\ 899--905. IEEE, 2013

    2013

  40. [40]

    R., and Leskovec, J

    Paranjape, A., Benson, A. R., and Leskovec, J. Motifs in temporal networks. In Proceedings of the tenth ACM international conference on web search and data mining, pp.\ 601--610, 2017

    2017

  41. [41]

    and Chen, Y

    Paul, S. and Chen, Y. Spectral and matrix factorization methods for consistent community detection in multi-layer networks. The Annals of Statistics, 48 0 (1): 0 230--250, 2020. doi:10.1214/18-AOS1800

    work page doi:10.1214/18-aos1800 2020

  42. [42]

    and Scherer, G

    Pereyra, V. and Scherer, G. Efficient computer manipulation of tensor products with applications to multidimensional approximation. Math. Comp., 27: 0 595--605, 1973

    1973

  43. [43]

    Deepwalk: Online learning of social representations

    Perozzi, B., Al-Rfou, R., and Skiena, S. Deepwalk: Online learning of social representations. In Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.\ 701--710, 2014

    2014

  44. [44]

    Bridging nestedness and economic complexity in multilayer world trade networks

    Ren, Z.-M., Zeng, A., and Zhang, Y.-C. Bridging nestedness and economic complexity in multilayer world trade networks. Humanities and Social Sciences Communications, 7 0 (1): 0 1--8, 2020

    2020

  45. [45]

    and Magnani, M

    Rossi, L. and Magnani, M. Towards effective visual analytics on multiplex and multilayer networks. Chaos, Solitons & Fractals, 72: 0 68--76, 2015. ISSN 0960-0779. doi:https://doi.org/10.1016/j.chaos.2014.12.022. Multiplex Networks: Structure, Dynamics and Applications

    work page doi:10.1016/j.chaos.2014.12.022 2015

  46. [46]

    S., Thiagarajan, J

    Shanthamallu, U. S., Thiagarajan, J. J., Song, H., and Spanias, A. Gramme: Semisupervised learning using multilayered graph attention models. IEEE transactions on neural networks and learning systems, 31 0 (10): 0 3977--3988, 2019

    2019

  47. [47]

    Improved Deep Embeddings for Inferencing with Multi-Layered Networks

    Song, H. and Thiagarajan, J. J. Improved deep embeddings for inferencing with multi-layered networks. arXiv preprint arXiv:1811.12156, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  48. [48]

    Tang, W., Lu, Z., and Dhillon, I. S. Clustering with multiple graphs. In 2009 Ninth IEEE International Conference on Data Mining, pp.\ 1016--1021. IEEE, 2009

    2009

  49. [49]

    Tucker, L. R. Some mathematical notes on three-mode factor analysis. Psychometrika, 31 0 (3): 0 279--311, 1966. doi:10.1007/BF02289464

    work page doi:10.1007/bf02289464 1966

  50. [50]

    Van Den Oord, E. J. and Van Rossem, R. Differences in first graders' school adjustment: The role of classroom characteristics and social structure of the group. Journal of School Psychology, 40 0 (5): 0 371--394, 2002

    2002

  51. [51]

    Van der Vaart, A. W. Asymptotic statistics, volume 3. Cambridge university press, 2000

    2000

  52. [52]

    Identifying key nodes in multilayer networks based on tensor decomposition

    Wang, D., Wang, H., and Zou, X. Identifying key nodes in multilayer networks based on tensor decomposition. Chaos: An Interdisciplinary Journal of Nonlinear Science, 27 0 (6), 2017 a

    2017

  53. [53]

    Tensor networks meet neural networks: A survey and future perspectives

    Wang, M., Pan, Y., Xu, Z., Li, G., Yang, X., Mandic, D., and Cichocki, A. Tensor networks meet neural networks: A survey and future perspectives. arXiv preprint arXiv:2302.09019, 2023 a

    work page arXiv 2023

  54. [54]

    Federated knowledge graph completion via latent embedding sharing and tensor factorization

    Wang, M., Zeng, D., Xu, Z., Guo, R., and Zhao, X. Federated knowledge graph completion via latent embedding sharing and tensor factorization. In 2023 IEEE International Conference on Data Mining (ICDM), pp.\ 1361--1366. IEEE, 2023 b

    2023

  55. [55]

    Tensorized hypergraph neural networks

    Wang, M., Zhen, Y., Pan, Y., Zhao, Y., Zhuang, C., Xu, Z., Guo, R., and Zhao, X. Tensorized hypergraph neural networks. In Proceedings of the 2024 SIAM International Conference on Data Mining (SDM), pp.\ 127--135. SIAM, 2024

    2024

  56. [56]

    Community preserving network embedding

    Wang, X., Cui, P., Wang, J., Pei, J., Zhu, W., and Yang, S. Community preserving network embedding. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 31, 2017 b

    2017

  57. [57]

    J., and Ng, A

    Xing, E., Jordan, M., Russell, S. J., and Ng, A. Distance metric learning with application to clustering with side-information. Advances in neural information processing systems, 15, 2002

    2002

  58. [58]

    Attentional multi-graph convolutional network for regional economy prediction with open migration data

    Xu, F., Li, Y., and Xu, S. Attentional multi-graph convolutional network for regional economy prediction with open migration data. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp.\ 2225--2233, 2020

    2020

  59. [59]

    How Powerful are Graph Neural Networks?

    Xu, K., Hu, W., Leskovec, J., and Jegelka, S. How powerful are graph neural networks? arXiv preprint arXiv:1810.00826, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  60. [60]

    Multisage: Empowering gcn with contextualized multi-embeddings on web-scale multipartite networks

    Yang, C., Pal, A., Zhai, A., Pancha, N., Han, J., Rosenberg, C., and Leskovec, J. Multisage: Empowering gcn with contextualized multi-embeddings on web-scale multipartite networks. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pp.\ 2434--2443, 2020

    2020

  61. [61]

    Y., Medvedovic, M., and Bumgarner, R

    Yeung, K. Y., Medvedovic, M., and Bumgarner, R. E. Clustering gene-expression data with repeated measurements. Genome Biology, 4 0 (5): 0 R34, 2003. doi:10.1186/gb-2003-4-5-r34

    work page doi:10.1186/gb-2003-4-5-r34 2003

  62. [62]

    T., and Ranka, S

    Yousefzadeh, N., Thai, M. T., and Ranka, S. A comprehensive survey on multi-layer graph embedding methods. Vietnam Journal of Computer Science, pp.\ 1--40, 2025

    2025

  63. [63]

    and Qu, A

    Yuan, Y. and Qu, A. Community detection with dependent connectivity. The Annals of Statistics, 49 0 (4): 0 2378--2428, 2021

    2021

  64. [64]

    Scalable multiplex network embedding

    Zhang, H., Qiu, L., Yi, L., and Song, Y. Scalable multiplex network embedding. In IJCAI, volume 18, pp.\ 3082--3088, 2018

    2018

  65. [65]

    Tensor generalized estimating equations for longitudinal imaging analysis

    Zhang, X., Li, L., Zhou, H., Zhou, Y., Shen, D., et al. Tensor generalized estimating equations for longitudinal imaging analysis. Statistica Sinica, 29 0 (4): 0 1977, 2019

    1977

  66. [66]

    Control of multilayer biological networks and applied to target identification of complex diseases

    Zheng, W., Wang, D., and Zou, X. Control of multilayer biological networks and applied to target identification of complex diseases. BMC bioinformatics, 20: 0 1--12, 2019

    2019