Pith. sign in

REVIEW 3 major objections 7 minor 56 references

Tri-Learn Graph Fusion Network for Attributed Graph Clustering

T0 review · 3 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper claims a tri-learning fusion of GCN, autoencoder, and graph transformer branches beats thirteen baselines on seven attributed-graph benchmarks, gaining up to 14.14% accuracy over its strongest rival.

desk verdict A plausible triple-fusion clustering idea undermined by equations that cannot be executed; the claims need verification from the code. read the letter →

arxiv 2507.13620 v2 pith:GK5XTKJM submitted 2025-07-18 cs.LG

Binxiong Li , Xu Xiang , Xue Li , Binyu Zhao , Heyang Gao , Qinyu Zhao This is my paper
classification cs.LG
keywords attributedgraphclusteringself-supervisedlearningtransformerconvolutionalnetworkautoencoderdeepfeaturefusion
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

Tri-GFN is proposed to settle a specific problem in attributed-graph clustering: how to use node attributes, local graph topology, and long-range dependencies together without the over-smoothing that deep graph convolutions suffer from. The paper's central claim is that a tri-learning network that fuses a GCN channel, an autoencoder channel, and a graph transformer channel - injecting their representations into one another at every layer and fusing them under a structure-diffusion step - yields more discriminative cluster assignments than any single- or dual-channel model. The model trains without labels, combining reconstruction losses with two KL-divergence self-supervision terms that align soft cluster assignments, and reports the top or near-top accuracy, NMI, ARI, and F1 on all seven benchmark datasets, with its largest accuracy gain, 14.14% over the strongest dual-network baseline, on the Reuters news corpus. If the claims hold, the framework supplies a practical recipe for clustering unlabeled attributed graphs - citation networks, sensor records, digit images, and news text - in which the three information channels reinforce rather than dilute one another.

What carries the argument

The carrying mechanism is the tri-learning fusion loop, with three parts. First, inter-layer injection (Eqs. 11-14) feeds the autoencoder's layer-$\ell$ representation $H^{(\ell)}$ into the GCN and graph transformer layers as $\varepsilon H^{(\ell)} + (1-\varepsilon) Z^{(\ell)}$, so attribute-level features steer the graph operators at every depth. Second, triple-channel enhancement (Eq. 15) mixes the three final branch outputs as $\mathbf{Z}_L = \tilde{\mathbf{A}}(\lambda \mathbf{Z}_{\mathrm{GCN}} + \theta \mathbf{Z}_{\mathrm{AE}} + \gamma \mathbf{Z}_T)$ and diffuses the mixture over the normalized adjacency matrix, so the fused embedding carries local topology, attributes, and long-range structure at once. Third, dual self-supervision turns a Student's $t$ soft assignment $Q$ (Eq. 16) into a target distribution $P$ (Eq. 17), then minimizes $\mathrm{KL}(Q \parallel P)$ and $\mathrm{KL}(Q \parallel Q')$ (Eqs. 18-19) to keep the three branches aligned on a single clustering, while the total loss (Eq. 24) adds feature- and adjacency-reconstruction terms scaled by $\alpha$ and $\beta$.

What would settle it

Two checks would settle the central claim. The first is dimensional: with $E$ defined as a $2\times |E|$ edge-index list, the product $E Z$ in Eqs. (4), (6), (13), and (14) is not defined for a node-feature matrix $Z \in \mathbb{R}^{N \times d}$, so running the code the manuscript makes available and comparing its forward pass against the written equations would reveal which operator — normalized adjacency, attention, or something else — actually produced the reported tables. The second is reproducibility under the written recipe: the paper concedes in Section 5 that the fusion weights $\lambda$, $\theta$, $\gamma$ had to be hand-tuned per dataset, so re-running with fixed seeds and checking whether Reuters reproduces ACC 81.86±0.18, USPS 81.01±0.10, and ACM 93.80±0.12, and whether any single weight set matches the per-dataset optima of Table 5, would separate the method's contribution from per-dataset search.

Watch

Extended reading notes

Core claim

The paper's core claim is that three-way mutual learning among GCN, autoencoder, and graph transformer branches achieves top or runner-up scores against every single- and dual-network clustering method in its comparison set. The GCN channel aggregates local neighborhoods, the autoencoder preserves node attributes through reconstruction, and the graph transformer channel, built on TransformerConv, an attention-weighted message-passing operator, captures long-range dependencies; at each encoder layer the autoencoder representation is blended into the other two channels as $\varepsilon H^{(\ell)} + (1-\varepsilon) Z^{(\ell)}$ (Eqs. 11-14), and the three final branch outputs are combined as $\mathbf{Z}_L = \tilde{\mathbf{A}}(\lambda \mathbf{Z}_{\mathrm{GCN}} + \theta \mathbf{Z}_{\mathrm{AE}} + \gamma \mathbf{Z}_T)$ and diffused over the normalized adjacency matrix (Eq. 15). Dual self-supervision then computes a soft assignment $Q$ from the fused representation, a second assignment $Q'$ from the autoencoder alone, and a target distribution $P$ from $Q$, minimizing $\mathrm{KL}(Q \parallel P)$ and $\mathrm{KL}(Q \parallel Q')$ so the branches converge on one clustering, with the final label drawn from $Q'$ (Eq. 26). On the seven datasets the model reports mean accuracies of 93.80 (ACM), 78.99 (DBLP), 71.55 (Citeseer), 73.93 (Cora), 84.51 (HHAR), 81.86 (Reuters), and 81.01 (USPS), the best or runner-up score in every table.

Load-bearing premise

The load-bearing premise is that the written equations describe the model that ran: Eqs. (4), (6), (13), and (14) multiply the raw edge-index list $E \in \mathbb{R}^{2\times |E|}$ directly against the node-feature matrix, an operation that is dimensionally undefined and is neither the normalized-adjacency convolution of Eq. (5) nor the attention mechanism of Eqs. (7)-(9), so the reported tables rest on an implementation the text does not specify.

Editorial extensions

If this is right

  • If the reported numbers hold, the three-channel recipe transfers across heterogeneous attributed graphs — citation networks, sensor-activity records, handwritten digits, and news corpora — because the model reports top or near-top results on all seven datasets.
  • The Reuters result in particular, a 14.14% accuracy gain over the strongest dual-network baseline with NMI and ARI also up sharply, would make Tri-GFN a practical label-free tool for automatic news classification and topic retrieval, a use the paper explicitly claims.
  • The propagation-depth study selects a three-layer configuration (Tri-GFN-3), since four layers degrade most datasets through over-smoothing; the framework's value lies in channel width and mutual learning rather than stack depth.
  • The ablation identifies the autoencoder channel as the most load-bearing: removing it lowers mean accuracy from 0.812 to 0.726 across datasets, while removing the GCN or transformer channels costs less, consistent with the claim that attribute information anchors the representations the graph channels refine.

Reading between the lines

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

  • Editorial inference: the gain over the dual-network baseline is large on Reuters (14.14%) and USPS (7.58%) but small on the structure-rich ACM citation graph (0.87%), which suggests the transformer channel earns its keep where local neighborhoods are weak or noisy; a testable extension is to regress per-dataset gains on graph density or homophily.
  • Editorial inference: the per-dataset optimal weights of Table 5, together with the random-forest analysis finding $\theta$ and $\gamma$ nearly equally influential, imply there is no universal fusion weight; a learned weighting module or a metric-based predictor of the composite score could replace the manual per-dataset search the authors concede is time-consuming.
  • Editorial inference: because the final labels come from $Q'$, the autoencoder's own assignment, rather than from the fused $Q$ (Eq. 26), the branch that casts the deciding vote is also the branch whose removal hurts most, leaving the GCN and transformer channels' contribution to the final output an open question the paper does not directly test.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. The paper proposes Tri-GFN, an attributed graph clustering model that combines an autoencoder, a GCN, and a Graph Transformer through a triple-channel enhancement module and two KL-divergence self-supervised losses. The authors evaluate the model on seven benchmark datasets, compare it with 13 methods, and report accuracy improvements of approximately 0.87% on ACM, 14.14% on Reuters, and 7.58% on USPS. The manuscript also contains ablation studies, hyperparameter sensitivity analyses, propagation-layer analyses, and statistical significance tests.

Significance. If the architecture were fully specified and the results reproducible, the work would be a meaningful empirical contribution to deep graph clustering: combining three heterogeneous encoders with dual self-supervision is a plausible strategy for improving clustering discrimination and mitigating GCN over-smoothing. The paper provides a public code link, a seven-dataset evaluation, and a wide range of baselines, ablations, and significance tests, which are notable strengths. However, the method as written is not a well-defined algorithm because the central layer-update equations are dimensionally invalid and the loss definitions are ambiguous; as a result, the reported benchmark numbers cannot be verified against the submitted specification.

major comments (3)
  1. [Secs. 3.4-3.6, Eqs. (4), (6), (13), (14)] The edge-index matrix E is defined in the text as E = [s; t] in R^{2 x |E|}, while each node embedding Z is N x d. The products E Z_GCN^(l), E Z_T^(l), and the analogous expressions in Eqs. (13)-(14) are therefore undefined for N != |E|, and even in the exceptional case they produce a 2 x d matrix rather than an N x d node embedding. These equations are the only layer-wise updates supplied for the GCN and Graph Transformer encoders. Eq. (5) provides a valid normalized-neighborhood aggregation, but it is never substituted into the layer updates, and Eq. (15) applies the normalized adjacency only once after fusion. The Q/K/V attention equations in (7)-(9) are likewise not used in Eq. (14). Consequently, the manuscript does not specify a computable forward pass, and the numerical results in Tables 2-3 cannot be traced to the written model. This is not a cosmetic typo, since the same invalid operator is reused in the core GCN, Graph Transformer, and fusion paths.
  2. [Sec. 3.8, Eqs. (20)-(24)] The objective function is also ambiguous. Eq. (20) defines L_GFN = L_w + delta L_e, but Eq. (21) introduces loss_w and loss_e without relating them to L_w and L_e; Eq. (22) defines L_AE using H_Gi_hat and then sets H_Gi_hat = H_AE_hat, which is self-referential; Eq. (23) defines Z_G_hat as an average of GCN and Graph Transformer decoder outputs, while Eq. (21) uses Z_Gi_hat; and Eq. (3) reconstructs the raw features X while Eq. (22) compares an AE reconstruction to A~X. These inconsistencies make the training loss, and therefore the reported optimized results, unreproducible. At minimum, a single coherent set of reconstruction losses with consistent notation is required.
  3. [Secs. 4.6.1, 4.6.2, and 5, Tables 1, 5, 6] The benchmarks are obtained under per-dataset manual selection of lambda, theta, gamma, alpha, beta, epsilon, learning rate, and epoch count, and the paper's own Section 5 states that 'we need to manually search for the optimal parameter sets possible for each dataset.' Table 5 reports the 'highest composite indices' achieved by these tuned settings, which suggests that the tuning criterion is the ground-truth-labeled clustering performance itself. Because ACC, NMI, ARI, and F1 all require ground-truth labels, using them to select hyperparameters and then reporting the best result as the method's performance can inflate the apparent advantage over baselines. A validation-based or fixed hyperparameter protocol is needed to support the stated state-of-the-art claims.
minor comments (7)
  1. [Abstract] The phrase 'framework that comprising' should be 'framework comprising', and the sentence beginning 'This comprehensive framework significantly improves clustering performance' is a fragment that should be joined to the preceding sentence.
  2. [Sec. 3.5, Eqs. (6) and (14)] Eqs. (6) and (14) are described as self-attention operations, but they contain no attention coefficients; the Q/K/V attention machinery in Eqs. (7)-(9) is never plugged into the layer update, so the Graph Transformer module as written is identical in form to the invalid GCN update.
  3. [Fig. 1 and Sec. 3.6] Figure 1 refers to final outputs Z(4), H(4), and T(4), while the text defines encoder layers only up to L/2 and Eq. (15) takes representations with superscript L/2; the layer indexing should be made consistent.
  4. [Algorithm 1, lines 8-9] Line 8 says 'Calculate the soft assignments Q, Q' and P, Q by Eq. (18) & Eq. (19)' and line 9 then says 'Generate the target distribution P by Eq. (17)', which places Eq. (17) after Eq. (18) even though Eq. (18) uses P; the order should be Q, Q', then P, then the two KL losses.
  5. [Sec. 4.4] The paper reports 'PyTorch 3.9.0', which is not a valid PyTorch version; the authors presumably mean Python 3.9 with some PyTorch release, and the exact versions should be stated.
  6. [References] Several in-text citations have no corresponding reference-list entries, including DCN (Yang et al., 2017), Caron et al. (2018), Zhang et al. (2021), and Pascanu (2013).
  7. [Sec. 4.5, Table 4] The continuation of Table 4 is labeled 'Continued Table 3' on page 28; it should read 'Continued Table 4'.

Circularity Check

1 steps flagged · score 4.0 of 10

Target distribution P is computed from the model's own soft assignments Q, making the clustering loss self-referential by construction; the benchmark superiority claims are empirical and not forced by this loop.

  1. self definitional [Section 3.7, Eqs. (16)-(18), and text after Eq. (18)]
    "q_ij = (1+||z_i - c_j||^2/t)^(-(t+1)/2) / sum_j' (...)^(-(t+1)/2) (16); p_ij = (q_ij^2 / f_j) / sum_j' (q_ij'^2 / f_j') (17); L_clu = KL(P || Q) = sum_i sum_j p_ij log(p_ij / q_ij) (18). This loss function helps to optimize the neural network's ability to approximate data representations closer to cluster centers, embodying the essence of self-supervision: P is derived from Q and in turn supervises its update."

    The target distribution P is defined by Eq. (17) as a sharpened, frequency-normalized function of the current soft assignments Q, and the optimization objective L_clu in Eq. (18) minimizes KL(P || Q). Thus the supervision signal contains no information beyond the model's own current assignments: training Q to match P is a self-referential sharpening of Q rather than fitting to an externally grounded target. The paper explicitly states this loop ('P is derived from Q and in turn supervises its update'), and the final label rule in Eq. (26) reads off the same family of soft assignments. By construction, the clustering 'prediction' is a transformation of the distribution used to define its own target.

full rationale

The only step that reduces by construction is the self-supervised target: Eq. (17) defines P from Q and Eq. (18) minimizes KL(P||Q), so the clustering loss is a self-referential sharpening of the model's own assignments. This is a standard DEC-style mechanism and is explicitly acknowledged in the text; it does not by itself force the reported accuracy numbers, which come from running the full three-module model on external benchmarks. I therefore do not treat the benchmark claims as circular. The per-dataset manual selection of lambda, theta, gamma, alpha, beta, epsilon, learning rate, and epochs (Section 4.6.1, Tables 5-6; Section 5) is a real selection-bias and overfitting risk, but it is not a by-construction equivalence: the reported numbers are selected maxima, not quantities that follow algebraically from the fitted parameters. Separately, the forward-pass equations (4), (6), (13), and (14) multiply the edge-index matrix E in R^{2 x |E|} by node features, which is dimensionally undefined; that is a model-specification and correctness problem, not a circularity. No load-bearing self-citation chain or uniqueness theorem is invoked, so the central architectural claim retains independent content despite the self-referential clustering loss.

Assumptions & free parameters 11 free parameters · 6 assumptions · 0 invented entities

The paper introduces no physical entities. Its central claim rests on standard deep-clustering machinery and on a large set of per-dataset tuned hyperparameters. The equations do not fully determine the implementation, so a reader must accept the listed assumptions to trust the reported tables.

free parameters (11)
  • lambda (GCN fusion weight) = ACM 0.5, DBLP 0.3, Citeseer 0.3, Cora 0.1, HHAR 0.1, Reuters 0.5, USPS 0.2
    Weight for the GCN module in Eq. (15); tuned per dataset and reported as optimal in Table 5.
  • theta (AE fusion weight) = ACM 0.4, DBLP 0.4, Citeseer 0.5, Cora 0.4, HHAR 0.4, Reuters 0.1, USPS 0.3
    Weight for the AE module in Eq. (15); tuned per dataset in Table 5.
  • gamma (Graph Transformer fusion weight) = ACM 0.1, DBLP 0.3, Citeseer 0.2, Cora 0.5, HHAR 0.5, Reuters 0.4, USPS 0.5
    Weight for the Graph Transformer module in Eq. (15); tuned per dataset in Table 5.
  • epsilon (layer fusion coefficient) = Table 1 lists 0.5, 0.5, 0.3, 0.5, 0.5, 0.9, 0.5
    Controls AE-feature injection into GCN and Graph Transformer layers in Eqs. (11)-(14); the text says it is fixed at 0.5, but Table 1 lists exceptions, notably 0.3 for Citeseer and 0.9 for Reuters.
  • alpha (clustering loss weight) = ACM 0.12, DBLP 0.1, Citeseer 0.15, Cora 0.1, HHAR 0.3, Reuters 0.3, USPS 0.1
    Weight on L_clu in Eq. (24); optimal per dataset in Table 6.
  • beta (consistency loss weight) = ACM 0.1, DBLP 0.12, Citeseer 0.3, Cora 0.12, HHAR 0.1, Reuters 0.15, USPS 0.1
    Weight on L_con in Eq. (24); optimal per dataset in Table 6.
  • learning rate = 5e-5, 2e-3, 4e-5, 1e-4, 1e-4, 1e-4, 1e-3
    Per-dataset learning rate listed in Table 1.
  • number of training epochs = ACM 200, DBLP 200, Citeseer 200, Cora 400, HHAR 600, Reuters 200, USPS 400
    Per-dataset training epochs in Table 1.
  • latent dimension n_z = 10, 10, 10, 10, 20, 20, 10
    Final embedding dimension per dataset in Table 1.
  • delta (adjacency reconstruction weight) = not reported
    Appears in Eq. (20) as the balance between feature and adjacency reconstruction losses, but no value is ever assigned.
  • number of propagation layers = 3
    Selected after comparing Tri-GFN-1 through Tri-GFN-4 in Section 4.9; the choice is made from benchmark performance rather than derived.
assumptions (6)
  • standard math An undirected attributed graph can be represented by feature matrix X, edge-index matrix E, and normalized adjacency A~ with self-loops (Eq. 5).
    Used throughout Section 3 as the graph representation and diffusion operator.
  • domain assumption The number of clusters k is known in advance for every dataset.
    Algorithm 1 takes k as input; no model-based cluster-number selection is evaluated.
  • domain assumption KMEANS on pre-trained AE representations provides a good initialization for cluster centroids.
    Stated in Section 3.7; a poor initialization would change all downstream soft assignments and the final clustering.
  • domain assumption The cluster sum of squares internal validity index identifies the optimal cluster arrangement.
    Invoked in Section 3.7 with reference [38]; the paper does not validate this choice on the benchmarks.
  • domain assumption The sharpened target distribution P derived from Q in Eq. (17) is a valid self-supervision signal.
    This is the standard DEC assumption; because P is the model's own output, the clustering loss is self-referential.
  • domain assumption Baseline numbers cited from prior papers were measured under comparable conditions.
    Section 4.4 says results are referenced from DDGAE, MBN, and SDCN; no baseline code was rerun, and missing entries are marked as not converging.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tri-Learn Graph Fusion Network for Attributed Graph Clustering." pith.science (2026). https://pith.science/paper/GK5XTKJM

@misc{pith2026250713620,
  author       = {Pith},
  title        = {Pith review of: Tri-Learn Graph Fusion Network for Attributed Graph Clustering},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GK5XTKJM}},
  note         = {Machine review of arXiv:2507.13620}
}
read the original abstract

In recent years, models based on Graph Convolutional Networks (GCN) have made significant strides in the field of graph data analysis. However, challenges such as over-smoothing and over-compression remain when handling large-scale and complex graph datasets, leading to a decline in clustering quality. Although the Graph Transformer architecture has mitigated some of these issues, its performance is still limited when processing heterogeneous graph data. To address these challenges, this study proposes a novel deep clustering framework that comprising GCN, Autoencoder (AE), and Graph Transformer, termed the Tri-Learn Graph Fusion Network (Tri-GFN). This framework enhances the differentiation and consistency of global and local information through a unique tri-learning mechanism and feature fusion enhancement strategy. The framework integrates GCN, AE, and Graph Transformer modules. These components are meticulously fused by a triple-channel enhancement module, which maximizes the use of both node attributes and topological structures, ensuring robust clustering representation. The tri-learning mechanism allows mutual learning among these modules, while the feature fusion strategy enables the model to capture complex relationships, yielding highly discriminative representations for graph clustering. It surpasses many state-of-the-art methods, achieving an accuracy improvement of approximately 0.87% on the ACM dataset, 14.14 % on the Reuters dataset, and 7.58 % on the USPS dataset. Due to its outstanding performance on the Reuters dataset, Tri-GFN can be applied to automatic news classification, topic retrieval, and related fields.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

56 extracted references · 44 canonical work pages

  1. [1]

    LeCun, Y

    Y. LeCun, Y. Bengio, G. Hinton, Deep learning, Nature 521 (2015) 436–444

    2015

  2. [2]

    Y. Wang, X. Yu, X. Guo, X. Wang, Y. Wei, S. Zeng, A dual -decoding branch U -shaped semantic segmentation network combining transformer attention with decoder: DBUNet, Journal of Visual Communication and Image Representation 95 (2023) 103856

    work page 2023

  3. [3]

    J. Xie, R. Girshick, A. Farhadi, Unsupervised deep embedding for clustering analysis, in: International Conference on Machine Learning, PMLR, 2016: pp. 478 –487. Binxiong Li et al., Tri-Learn Graph Fusion Network for Attributed Graph Clustering 45 http://proceedings.mlr.press/v48/xieb16.html

    work page 2016

  4. [4]

    Jiang, Y

    Z. Jiang, Y. Zheng, H. Tan, B. Tang, H. Zhou, Variational Deep Embedding: An Unsupervised and Generative Approach to Clustering, (2017). http://arxiv.org/abs/1611.05148

    arXiv 2017

  5. [5]

    D. Hu, D. Feng, Y. Xie, EGC: A novel event-oriented graph clustering framework for social media text, Information Processing & Management 59 (2022) 103059

    work page 2022

  6. [6]

    Velickovic, W

    P. Velickovic, W. Fedus, W.L. Hamilton, P. Liò, Y. Bengio, R.D. Hjelm, Deep graph infomax., ICLR (Poster) 2 (2019) 4

    work page 2019

  7. [7]

    T.N. Kipf, M. Welling, Variational Graph Auto -Encoders, (2016). http://arxiv.org/abs/1611.07308

    arXiv 2016

  8. [8]

    D. Bo, X. Wang, C. Shi, M. Zhu, E. Lu, P. Cui, Structural Deep Clustering Network, in: Proceedings of The Web Conference 2020, ACM, Taipei Taiwan, 2020: pp. 1400 –1410. https://doi.org/10.1145/3366423.3380214

    arXiv 2020

Show all 56 references
  1. [9]

    Zhang, H

    X. Zhang, H. Liu, Q. Li, X. -M. Wu, X. Zhang, Adaptive Graph Convolution Methods for Attributed Graph Clustering, IEEE Transactions on Knowledge and Data Engineering 35 (2023) 12384–12399

    2023

  2. [10]

    Zhang, H

    X. Zhang, H. Liu, Q. Li, X. -M. Wu, Attributed Graph Clustering via Adaptive Graph Convolution, (2019). http://arxiv.org/abs/1906.01210

    2019 arXiv

  3. [11]

    T.N. Kipf, M. Welling, Semi-Supervised Classification with Graph Convolutional Networks, (2017). http://arxiv.org/abs/1609.02907

    2017 arXiv

  4. [12]

    X. Guo, L. Gao, X. Liu, J. Yin, Improved deep embedded clustering with local structure preservation., in: Ijcai, 2017: pp. 1753–1759

    2017

  5. [13]

    Z. Xie, M. Wu, G. Zhao, L. Zhou, Z. Gong, Z. Zhang, Attributed Multi -relational Graph Embedding Based on GCN, Lecture Notes in Computer Science (2023) 174 –186. https://doi.org/10.1007/978-981-99-4742-3_14

    2023 doi

  6. [14]

    S. Hu, G. Zou, C. Zhang, Z. Lou, R. Geng, Y. Ye, Joint contrastive triple -learning for deep multi-view clustering, Information Processing & Management 60 (2023) 103284

    2023

  7. [15]

    Newman, Modularity and community structure in networks, Proc

    M.E.J. Newman, Modularity and community structure in networks, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 8577–8582. https://doi.org/10.1073/pnas.0601602103

    2006 doi

  8. [16]

    Q. Li, Z. Han, X. -M. Wu, Deeper insights into graph convolutional networks for semi - supervised learning, in: Proceedings of the AAAI Conference on Artificial Intelligence, 2018. https://ojs.aaai.org/index.php/AAAI/article/view/11604

    2018

  9. [17]

    J. MacQueen, Some methods for classification and analysis of multivariate observations, in: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Oakland, CA, USA, 1967: pp. 281–297

    1967

  10. [18]

    Ester, H.-P

    M. Ester, H.-P. Kriegel, J. Sander, X. Xu, A density-based algorithm for discovering clusters in large spatial databases with noise, in: Kdd, 1996: pp. 226–231

    1996

  11. [19]

    Rumelhart, G.E

    D.E. Rumelhart, G.E. Hinton, R.J. Williams, Learning representations by back -propagating Pattern Analysis and Applications 46 errors, Nature 323 (1986) 533–536

    1986

  12. [20]

    Kingma, M

    D.P. Kingma, M. Welling, Auto -Encoding Variational Bayes, (2022). http://arxiv.org/abs/1312.6114

    2022 arXiv

  13. [21]

    Goodfellow, J

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, Y. Bengio, Generative adversarial nets, Advances in Neural Information Processing Systems 27 (2014)

    2014

  14. [22]

    Scarselli, M

    F. Scarselli, M. Gori, A.C. Tsoi, M. Hagenbuchner, G. Monfardini, The graph neural network model, IEEE Transactions on Neural Networks 20 (2008) 61–80

    2008

  15. [23]

    W. Tu, S. Zhou, X. Liu, X. Guo, Z. Cai, E. Zhu, J. Cheng, Deep fusion clustering network, in: Proceedings of the AAAI Conference on Artificial Intelligence, 2021: pp. 9978 –9987. https://ojs.aaai.org/index.php/AAAI/article/view/17198

    2021

  16. [24]

    J. Zhao, J. Guo, Y. Sun, J. Gao, S. Wang, B. Yin, Adaptive graph convolutional clustering network with optimal probabilistic graph, Neural Networks 156 (2022) 271–284

    2022

  17. [25]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is all you need, Advances in Neural Information Processing Systems 30 (2017)

    2017

  18. [26]

    J. Lee, I. Lee, J. Kang, Self-Attention Graph Pooling, (2019). http://arxiv.org/abs/1904.08082 (accessed September 17, 2024)

    2019 arXiv

  19. [27]

    Velickovic, G

    P. Velickovic, G. Cucurull, A. Casanova, A. Romero, P. Lio, Y. Bengio, Graph attention networks, Stat 1050 (2017) 10–48550

    2017

  20. [28]

    G. Huo, Y. Zhang, J. Gao, B. Wang, Y. Hu, B. Yin, CaEGCN: Cross -attention fusion based enhanced graph convolutional network for clustering, IEEE Transactions on Knowledge and Data Engineering 35 (2021) 3471–3483

    2021

  21. [29]

    Hoang, O.-J

    V.T. Hoang, O.-J. Lee, Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role -based Similarity, (2023). http://arxiv.org/abs/2308.09517 (accessed September 17, 2024)

    2023 arXiv

  22. [30]

    Zhang, Y

    C. Zhang, Y. Zhao, J. Wang, Transformer -based dynamic fusion clustering network, Knowledge-Based Systems 258 (2022) 109984

    2022

  23. [31]

    S.J. Ahn, M. Kim, Variational Graph Normalized AutoEncoders, in: Proceedings of the 30th ACM International Conference on Information & Knowledge Management, ACM, Virtual Event Queensland Australia, 2021. https://doi.org/10.1145/3459637.3482215

    2021

  24. [32]

    Manchanda, D

    S. Manchanda, D. Zheng, G. Karypis, Schema-aware deep graph convolutional networks for heterogeneous graphs, in: 2021 IEEE International Conference on Big Data (Big Data), IEEE, 2021: pp. 480–489. https://ieeexplore.ieee.org/abstract/document/9671380/

    2021

  25. [33]

    Huang, Z

    J. Huang, Z. Li, N. Li, S. Liu, G. Li, Attpool: Towards hierarchical feature representation in graph convolutional networks via attention mechanism, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: pp. 6480 –6489. http://openaccess.thecvf.com/...

    2019

  26. [34]

    X. Yan, X. Yu, S. Hu, Y. Ye, Mutual Boost Network for attributed graph clustering, Expert Systems with Applications 229 (2023) 120479

    2023

  27. [35]

    Z.-P. Li, H. -L. Su, X. -B. Zhu, X. -M. Wei, X. -S. Jiang, V. Gribova, V.F. Filaretov, D. -S. Huang, Hierarchical graph pooling with self-adaptive cluster aggregation, IEEE Transactions on Cognitive and Developmental Systems 14 (2021) 1198–1207

    2021

  28. [36]

    Y. Shi, Z. Huang, S. Feng, H. Zhong, W. Wang, Y. Sun, Masked Label Prediction: Unified Message Passing Model for Semi -Supervised Classification, (2021). http://arxiv.org/abs/2009.03509

    2021 arXiv

  29. [37]

    Van der Maaten, G

    L. Van der Maaten, G. Hinton, Visualizing data using t -SNE., Journal of Machine Learning Research 9 (2008)

    2008

  30. [38]

    R. Ünlü, P. Xanthopoulos, Estimating the number of clusters in a dataset via consensus clustering, Expert Systems with Applications 125 (2019) 33–39

    2019

  31. [39]

    Ley, DBLP: some lessons learned, Proc

    M. Ley, DBLP: some lessons learned, Proc. VLDB Endow. 2 (2009) 1493 –1500. https://doi.org/10.14778/1687553.1687577

    2009

  32. [40]

    Giles, K.D

    C.L. Giles, K.D. Bollacker, S. Lawrence, CiteSeer: an automatic citation indexing system, in: Proceedings of the Third ACM Conference on Digital Libraries - DL ’98, ACM Press, Pittsburgh, Pennsylvania, United States, 1998: pp. 89 –98. https://doi.org/10.1145/276675.276685

    1998

  33. [41]

    Stisen, H

    A. Stisen, H. Blunck, S. Bhattacharya, T.S. Prentow, M.B. Kjærgaard, A. Dey, T. Sonne, M.M. Jensen, Smart Devices are Different: Assessing and MitigatingMobile Sensing Heterogeneities for Activity Recognition, in: Proceedings of the 13th ACM Conference on Embedded Networked Se...

    2015

  34. [42]

    McCallum, K

    A.K. McCallum, K. Nigam, J. Rennie, K. Seymore, Automating the Construction of Internet Portals with Machine Learning, (n.d.)

  35. [43]

    LeCun, B

    Y. LeCun, B. Boser, J. Denker, D. Henderson, R. Howard, W. Hubbard, L. Jackel, Handwritten digit recognition with a back -propagation network, Advances in Neural Information Processing Systems 2 (1989)

    1989

  36. [44]

    Padmanabhan, S

    D. Padmanabhan, S. Bhat, S. Shevade, Y. Narahari, Topic model based multi -label classification, in: 2016 IEEE 28th International Conference on Tools with Artificial Intelligence (ICTAI), IEEE, 2016: pp. 996 –1003. https://ieeexplore.ieee.org/abstract/document/7814714/

    2016

  37. [45]

    N.X. Vinh, J. Epps, J. Bailey, Information Theoretic Measures for Clusterings Comparison: Variants, Properties, Normalization and Correction for Chance 18 (2009)

    2009

  38. [46]

    Chinchor, B.M

    N. Chinchor, B.M. Sundheim, MUC-5 evaluation metrics, in: Fifth Message Understanding Conference (MUC-5): Proceedings of a Conference Held in Baltimore, Maryland, August 25- Pattern Analysis and Applications 48 27, 1993, 1993. https://aclanthology.org/M93-1007.pdf

    1993

  39. [47]

    C. Wang, S. Pan, R. Hu, G. Long, J. Jiang, C. Zhang, Attributed Graph Clustering: A Deep Attentional Embedding Approach, (2019). http://arxiv.org/abs/1906.06532

    2019 arXiv

  40. [48]

    X. Wu, W. Lu, Y. Quan, Q. Miao, P.G. Sun, Deep dual graph attention auto -encoder for community detection, Expert Systems with Applications 238 (2024) 122182

    2024

  41. [49]

    Y. Liu, W. Tu, S. Zhou, X. Liu, L. Song, X. Yang, E. Zhu, Deep graph clustering via dual correlation reduction, in: Proceedings of the AAAI Conference on Artificial Intelligence, 2022: pp. 7603–7611. https://ojs.aaai.org/index.php/AAAI/article/view/20726

    2022

  42. [50]

    Breiman, RANDOM FORESTS, (2001)

    L. Breiman, RANDOM FORESTS, (2001). http://wwww.saedsayad.com/docs/rf.pdf

    2001

  43. [51]

    Elston, On Fisher’s Method of Combining p‐Values, Biometrical J 33 (1991) 339–345

    R.C. Elston, On Fisher’s Method of Combining p‐Values, Biometrical J 33 (1991) 339–345. https://doi.org/10.1002/bimj.4710330314

    1991 doi

  44. [52]

    Jomaa, L

    H.S. Jomaa, L. Schmidt -Thieme, J. Grabocka, Hyperparameter Optimization with Differentiable Metafeatures, (2021). http://arxiv.org/abs/2102.03776

    2021 arXiv

  45. [53]

    Bergstra, Y

    J. Bergstra, Y. Bengio, Random search for hyper-parameter optimization., Journal of Machine Learning Research 13 (2012). https://www.jmlr.org/papers/volume13/bergstra12a/bergstra12a.pdf?ref=broutonlab.com

    2012

  46. [54]

    Yperman, T

    J. Yperman, T. Becker, Bayesian optimization of hyper -parameters in reservoir computing, (2017). http://arxiv.org/abs/1611.05193

    2017 arXiv

  47. [55]

    Parsa, A

    M. Parsa, A. Ankit, A. Ziabari, K. Roy, Pabo: Pseudo agent -based multi-objective bayesian hyperparameter optimization for efficient neural accelerator design, in: 2019 IEEE/ACM International Conference on Computer -Aided Design (ICCAD), IEEE, 2019: pp. 1 –8. https://ieeexplor...

    2019

  48. [56]

    Javad Shafiee, P

    M. Javad Shafiee, P. Siva, P. Fieguth, A. Wong, Efficient deep feature learning and extraction via stochasticnets, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2016: pp. 28–36

    2016

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.