Pith. sign in

REVIEW 4 major objections 6 minor 39 references

Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning

T0 review · 4 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A symmetric graph convolutional autoencoder reconstructs node features by pairing Laplacian smoothing in the encoder with numerically stable Laplacian sharpening in the decoder.

desk verdict The GALA decoder is a real architectural novelty with strong empirical results, but the paper's own claim that the stabilized operator preserves Laplacian sharpening is mathematically wrong for high-degree nodes and needs fixing. read the letter →

arxiv 1908.02441 v1 pith:GWZVJVCO submitted 2019-08-07 cs.LG cs.CVstat.ML

Jiwoong Park , Minsik Lee , Hyung Jin Chang , Kyuewang Lee , Jin Young Choi This is my paper · ORCID
classification cs.LGcs.CVstat.ML
keywords graphconvolutionalautoencoderLaplaciansmoothingsharpeningsignedgraphsunsupervisedrepresentationlearningnodeclusteringlinkpredictionsubspace
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

This paper introduces a graph convolutional autoencoder, named GALA, whose architecture is symmetric in a functional sense: the encoder applies Laplacian smoothing, pulling each node's latent representation toward its neighbors, and the decoder applies Laplacian sharpening, pushing reconstructed features away from the neighbor centroid. Because the decoder uses the same graph structure as the encoder and reconstructs node features directly, the model avoids the asymmetric graph autoencoders whose decoders reconstruct only the affinity matrix or perform a linear mapping. The straightforward sharpening operator has spectral radius 3, so the paper rewrites it using a signed graph with self-loop weight 2, obtaining $\hat{D}^{-1/2}\hat{A}\hat{D}^{-1/2}$ with $\hat{A}=2I_n-A$ and $\hat{D}=2I_n+D$, whose spectral radius is 1, and argues this stabilized operator preserves the meaning of the original sharpening. A subspace-clustering cost derived from least-squares regression is added for image clustering, yielding a low-dimensional latent affinity matrix together with the latent representation. On the reported benchmarks, this yields gains over earlier graph autoencoders on unsupervised node clustering and link prediction.

What carries the argument

The central object is the numerically stable Laplacian sharpening operator $\hat{D}^{-1/2}\hat{A}\hat{D}^{-1/2}$, where $\hat{A}=2I_n - A$ and $\hat{D}=2I_n + D$; it is the decoder propagation of Eq. (16). The paper obtains it from the signed graph induced by flipping all edge weights to negative and adding self-loops of weight 2, and relies on the theorem that the normalized Laplacian of a signed graph has eigenvalues in $[0,2]$, so the operator has spectral radius 1. It is the exact counterpart of the GCN encoder's Laplacian smoothing $\tilde{D}^{-1/2}\tilde{A}\tilde{D}^{-1/2}$ and is derived by setting $K=1$, $\lambda_{\max}\approx 2$ in a Chebyshev approximation of sharpening. The other load-bearing component is the subspace-clustering cost, minimized analytically as $\mu\lambda/2 \operatorname{tr}((\mu I_k+\lambda HH^T)^{-1}HH^T)$, which reduces matrix inversion from $n\times n$ to $k\times k$.

What would settle it

Take a star graph with a hub of degree $d$ and compute the stabilized operator's output on a one-hot feature at the hub; compare it with the ideal operator's output. If the hub's reconstructed value moves toward the average of its leaves rather than away from it, the stabilized operator is not sharpening for that node, and the claimed equivalence between Eqs. (14) and (15) fails.

Watch

Extended reading notes

Core claim

The central claim is that a completely symmetric graph convolutional autoencoder is possible and beneficial: the encoder is the renormalized GCN propagation $\tilde{D}^{-1/2}\tilde{A}\tilde{D}^{-1/2}$ of [14], a Laplacian smoothing, while the decoder is a newly derived Laplacian sharpening $\hat{D}^{-1/2}\hat{A}\hat{D}^{-1/2}$ with $\hat{A}=2I_n-A$ and $\hat{D}=2I_n+D$, built from a signed graph so that its spectral radius is 1. The paper claims this decoder reconstructs the node feature matrix $X$ directly and uses graph structure throughout, in contrast to existing graph autoencoders that reconstruct the affinity matrix or have non-learnable decoder parts. It further claims that the stabilized sharpening preserves the elementwise meaning of the ideal operator $2I_n - D^{-1/2}AD^{-1/2}$, and that adding the analytic subspace-clustering cost (Eq. 21) improves image clustering by finding a latent representation and a latent affinity matrix simultaneously. The reported experiments on citation networks and image datasets show gains over GAE/VGAE, MGAE, and ARGA/ARVGA on clustering and link prediction.

Load-bearing premise

The load-bearing premise is that the stabilized decoder $\hat{D}^{-1/2}\hat{A}\hat{D}^{-1/2}$ with $\hat{A}=2I_n - A$ and $\hat{D}=2I_n + D$ preserves the meaning of the ideal sharpening operator $2I_n - D^{-1/2}AD^{-1/2}$ for nodes of every degree; the paper asserts this equivalence without analyzing how the degree-dependent self-term $2/(D_{ii}+2)$ changes the operator's behavior on high-degree nodes.

Editorial extensions

If this is right

  • A decoder that sharpens while reconstructing node features should make the learned latent representations more clusterable, since the encoder's smoothing and decoder's sharpening jointly exploit graph structure rather than only the encoder doing so.
  • The spectral-radius-1 sharpening operator can be stacked across many decoder layers without the instability of repeated $2I_n - D^{-1/2}AD^{-1/2}$, so deeper symmetric graph autoencoders become feasible.
  • The analytic subspace-clustering cost turns an $O(n^3)$ problem into $O(k^3)$, making the added clustering objective practical whenever the latent dimension is small.
  • If the benchmark numbers hold, the model establishes a new default: feature-reconstruction graph autoencoders can surpass affinity-reconstruction and linear-mapping autoencoders on unsupervised node clustering and link prediction.

Reading between the lines

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

  • The stabilized operator's self-term is $2/(D_{ii}+2)$, so for high-degree nodes the self-weight is weaker relative to neighbor weights than in the ideal operator; the paper asserts equivalence without proving that sharpening survives on hubs, so one test is whether hub reconstructions actually move away from neighbor centroids.
  • The same signed-graph construction could give stable higher-order graph filters or stable decoders for other graph autoencoder designs, since only the spectral-radius bound is used.
  • If the smoothing/sharpening symmetry is the cause of the reported gains, then breaking the symmetry (e.g., by using different propagation orders or adding skip connections between encoder and decoder) should reduce clustering performance; that experiment is not in the paper.
  • The subspace-clustering cost is derived from least-squares regression and could be attached to any low-dimensional latent code, not just the GALA encoder, as a generic clustering regularizer.
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

4 major / 6 minor

Summary. The paper proposes GALA, a graph convolutional autoencoder with a symmetric encoder-decoder structure. The encoder applies renormalized Laplacian smoothing (Eq. 17), and the decoder applies a numerically stabilized operator derived from Laplacian sharpening (Eq. 16), constructed by taking A_hat = 2I - A and D_hat = 2I + D in a signed-graph framework. For image clustering, the authors add a least-squares subspace clustering cost and derive a reduced k-by-k formulation (Eq. 21). Experiments on node clustering, image clustering, link prediction, and visualization report strong results against VGAE, MGAE, ARGA, and ARVGA, with ablation studies supporting both the stabilized decoder and the subspace clustering cost.

Significance. If the theoretical claims are established, the paper makes a useful contribution: a graph decoder that uses both structure and node features, a stable variant of Laplacian sharpening, and a computationally efficient subspace clustering cost. The reported empirical gains are substantial (e.g., Cora ACC 0.7459 vs MGAE 0.6844; Citeseer link prediction AUC 94.4 vs ARVGA 92.4). The ablation design cleanly separates the effects of the decoder and the subspace clustering cost, and the derivation of the subspace cost from the analytic LSR solution is a credible reduction from O(n^3) to O(k^3). However, the central theoretical justification of the stabilized decoder, and several reproducibility details, need to be strengthened before the claims can be accepted.

major comments (4)
  1. [Section 3.2, Eqs. (14)-(16)] The claim that Eq. (14) 'has the same meaning' as Eq. (15) is not supported by the text. For a d-regular graph, the stable operator B = (2I+D)^{-1/2}(2I-A)(2I+D)^{-1/2} equals (d/(d+2))C - (2(d-1)/(d+2))I, where C = 2I - D^{-1/2}AD^{-1/2} is the ideal sharpening operator of Eq. (12). The all-ones eigenvalue of B is (2-d)/(d+2), which is negative for d>2 and tends to -1 as d grows, while the all-ones eigenvalue of C is 1. Thus for high-degree nodes the stabilized operator attenuates or inverts the low-frequency response instead of sharpening it; a learned weight matrix applied after B cannot repair a degree-dependent sign flip with a global linear map. The authors need to either prove a precise sense in which Eq. (16) preserves Laplacian sharpening, or reinterpret the decoder and the Table 1 ablation accordingly.
  2. [Section 3.2, Eq. (13)] The numerical-stability assertion is attributed to Theorem 1 of [18], but that theorem concerns normalized Laplacians of signed graphs whose adjacency matrices have zero diagonal, whereas A_hat = 2I - A has diagonal entries 2. The hypotheses of the cited theorem are not verified in the text. The spectral-radius bound may in fact hold for this specific construction, since both D_hat - A_hat = D + A and D_hat + A_hat = 4I + (D - A) are positive semidefinite, which implies |x^T A_hat x| <= x^T D_hat x for every x; however, this proof is not given, and the citation as written is not sufficient.
  3. [Sections 4.1-4.2 and 3.3] The implementation details necessary to reproduce the experiments are not in the manuscript. Table 3 gives no affinity matrix for COIL20, YALE, or MNIST, yet Eqs. (16) and (17) require a graph A; the construction of A for image data (for example, number of nearest neighbors and kernel parameters) is not stated. In addition, the derivation of Eq. (21), the hyperparameter settings, and the training details are deferred to a supplementary that is not included in this submission. These materials should be supplied in a revised version.
  4. [Sections 4.4-4.6, Tables 2 and 4-6] The clustering results are reported as means over 50 runs without standard deviations or other measures of dispersion. Some of the differences used to justify the method are small, such as GALA vs GALA+SCC on MNIST (ACC 0.7384 vs 0.7426; ARI 0.6469 vs 0.6675), and the ablation conclusions in Table 6 rely on such margins. The authors should report variances or significance tests to support the claims that the stable decoder and the subspace clustering cost each improve performance.
minor comments (6)
  1. [Section 4.5, Table 4] The sentence that GALA 'outperforms several baselines and the state-of-the-art graph convolution algorithms for most of the cases' is somewhat overstated: on YALE, GALA without the subspace cost achieves ACC 0.8530, below VGAE (0.9157), ARGA (0.9309), and ARVGA (0.8727). The strong YALE result comes from GALA+SCC, so the text should qualify the claim.
  2. [Section 4.6, Table 6] Similarly, the statement that the stable decoder with reconstruction cost only 'outperforms the state-of-the-art algorithms in most cases' is not true on YALE, where the stable reconstruction-only configuration gives ACC 0.8530, lower than several comparison methods in Table 4.
  3. [Notation, Eqs. (14) and (22)] The symbol A_hat is used both for the stabilized decoder adjacency A_hat = 2I - A in Section 3.2 and for the reconstructed affinity A_hat = sigmoid(HH^T) in the link prediction cost of Eq. (22). Using different symbols would reduce confusion.
  4. [Section 3.4, Eq. (18)] Eq. (18) writes the minimization over \bar X alone, but \bar X is the output of the decoder and depends on the network parameters; the text should state explicitly that the minimization is over the encoder and decoder parameters.
  5. [Section 4.8, Figure 3] The text refers to a red dotted line in the YALE embeddings, but the figure caption does not explain what the line marks; please add an explanation or remove the reference.
  6. [Section 3.1] The phrase 'completely symmetric autoencoder' should be defined more carefully: the encoder and decoder are not exact inverses, and 'symmetric' appears to mean that graph convolution is used in both stages. A brief definition would prevent misunderstanding.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the decoder and cost derivations are analytic or rest on external cited results, and empirical gains are benchmarked against external baselines.

full rationale

The paper's derivation chain is not circular. The encoder is the standard GCN renormalized convolution (Eqs. 7 and 17), taken from Kipf & Welling, and the decoder is derived by writing Laplacian sharpening in Chebyshev form (Eq. 12) and then renormalizing via a signed-graph Laplacian whose eigenvalue bound is cited to an external theorem (Li & Li, [18]). Nothing in this chain defines a prediction in terms of the target output or fits a parameter to labels it later claims to predict. The subspace-clustering cost follows from an exact analytic minimization of the LSR objective over the affinity variable A_H, so it is not a fitted input disguised as a prediction. The reported clustering and link-prediction numbers are benchmarked against external baselines on public datasets, and no cluster label is used during training. The skeptical concern about Eqs. (14) and (15) — that the diagonal entry 2/(D_ii+2) does not actually preserve the behavior of 2I - D^{-1/2}AD^{-1/2} for high-degree nodes — is a correctness or hypothesis-verification gap in the application of the signed-graph theorem, not a circularity: the paper does not reduce Eq. (16) to Eq. (12) by construction; it merely asserts 'same meaning.' Since no load-bearing step is equivalent to its own input by definition or through a fitted-value renaming, the circularity score is 0.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central decoder and cost function rest on a cited spectral theorem about signed graphs and standard spectral graph theory, plus three problem-specific hyperparameters (lambda, mu, gamma) and design choices (self-loop weight 2, network depth) whose values are deferred to the supplementary. No new physical or conceptual entities are introduced.

free parameters (5)
  • Decoder self-loop weight (2 in A_hat = 2I - A) = 2 (fixed by design)
    Chosen to match the ideal sharpening operator's diagonal; controls the balance between a node's own value and its neighbors in every decoder layer.
  • Subspace clustering weight lambda (Eqs. 19, 21) = not stated in main text
    Regularization for the self-expressiveness term; values deferred to supplementary material.
  • Subspace clustering weight mu (Eqs. 19, 21) = not stated in main text
    Regularization for the learned affinity matrix; values deferred to supplementary material.
  • Link prediction cost weight gamma (Eq. 22) = not stated in main text
    Balances feature reconstruction against link prediction in the Citeseer link prediction experiments.
  • Network depth M and hidden dimensions = not stated in main text
    Number of symmetric encoder/decoder layers and layer widths; deferred to supplementary material.
assumptions (4)
  • standard math Normalized Laplacian eigenvalues of signed graphs lie in [0,2] (Theorem 1 of [18]).
    This cited theorem is the basis for the claim that the decoder operator in Eq. (16) has spectral radius at most 1. The paper does not re-prove it.
  • standard math GCN renormalization trick (adding self-loops and renormalizing) preserves the meaning of Laplacian smoothing and is numerically stable.
    The encoder directly uses the renormalized GCN layer from [14]; stability is inherited from that prior analysis.
  • standard math Chebyshev truncation with K=1 and lambda_max approx 2 yields valid first-order graph spectral filters.
    Both encoder and decoder are justified as K=1 Chebyshev approximations following [10,14]; used in Eqs. (3), (12), and (16).
  • domain assumption Image data lie in multiple low-dimensional subspaces (self-expressiveness property).
    Motivates the subspace clustering cost in Eq. (19) for image clustering; without this assumption the added cost has no justification.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning." pith.science (2026). https://pith.science/paper/GWZVJVCO

@misc{pith2026190802441,
  author       = {Pith},
  title        = {Pith review of: Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GWZVJVCO}},
  note         = {Machine review of arXiv:1908.02441}
}
read the original abstract

We propose a symmetric graph convolutional autoencoder which produces a low-dimensional latent representation from a graph. In contrast to the existing graph autoencoders with asymmetric decoder parts, the proposed autoencoder has a newly designed decoder which builds a completely symmetric autoencoder form. For the reconstruction of node features, the decoder is designed based on Laplacian sharpening as the counterpart of Laplacian smoothing of the encoder, which allows utilizing the graph structure in the whole processes of the proposed autoencoder architecture. In order to prevent the numerical instability of the network caused by the Laplacian sharpening introduction, we further propose a new numerically stable form of the Laplacian sharpening by incorporating the signed graphs. In addition, a new cost function which finds a latent representation and a latent affinity matrix simultaneously is devised to boost the performance of image clustering tasks. The experimental results on clustering, link prediction and visualization tasks strongly support that the proposed model is stable and outperforms various state-of-the-art algorithms.

Figures

Figures reproduced from arXiv: 1908.02441 by the authors.

Figure 1
Figure 1. Architectures of existing graph convolutional au [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Sample images of three image datasets where X¯ is the reconstructed feature matrix of nodes, the column of X¯ corresponds to the output of the decoder for an input feature of a node, and k · kF denotes the Frobenius norm. 3.3. Subspace clustering cost for image clustering It is a well-known assumption that image datasets are often drawn from multiple low-dimensional subspaces, al￾though their data dimensions are hig… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 36 canonical work pages

  1. [18]

    Note on the normalized laplacian eigenvalues of signed graphs

    Hong Hai Li and Jiong Sheng Li. Note on the normalized laplacian eigenvalues of signed graphs. Australas. J. Com- bin, 44:153–162, 2009. 2, 4

    work page 2009

  2. [1]

    Laplacian eigenmaps and spectral techniques for embedding and clustering

    Mikhail Belkin and Partha Niyogi. Laplacian eigenmaps and spectral techniques for embedding and clustering. In Advances in Neural Information Processing Systems , pages 585–591, 2002. 1

    work page 2002

  3. [2]

    Geometric deep learning: going beyond euclidean data

    Michael M Bronstein, Joan Bruna, Yann LeCun, Arthur Szlam, and Pierre Vandergheynst. Geometric deep learning: going beyond euclidean data. IEEE Signal Processing Mag- azine, 34(4):18–42, 2017. 1

    2017

  4. [3]

    Deep neural net- works for learning graph representations

    Shaosheng Cao, Wei Lu, and Qiongkai Xu. Deep neural net- works for learning graph representations. In AAAI, pages 1145–1152, 2016. 6

    work page 2016

  5. [4]

    Relational topic models for document networks

    Jonathan Chang and David Blei. Relational topic models for document networks. In Artificial Intelligence and Statistics, pages 81–88, 2009. 6

    work page 2009

  6. [5]

    Spectral graph the- ory

    Fan RK Chung and Fan Chung Graham. Spectral graph the- ory. Number 92. American Mathematical Soc., 1997. 2

    work page 1997

  7. [6]

    Convolutional neural networks on graphs with fast localized spectral filtering

    Micha ¨el Defferrard, Xavier Bresson, and Pierre Van- dergheynst. Convolutional neural networks on graphs with fast localized spectral filtering. In Advances in Neural Infor- mation Processing Systems, pages 3844–3852, 2016. 2

    work page 2016

  8. [7]

    Convolutional networks on graphs for learn- ing molecular fingerprints

    David K Duvenaud, Dougal Maclaurin, Jorge Iparraguirre, Rafael Bombarell, Timothy Hirzel, Al´an Aspuru-Guzik, and Ryan P Adams. Convolutional networks on graphs for learn- ing molecular fingerprints. In Advances in Neural Informa- tion Processing Systems, pages 2224–2232, 2015. 1

    work page 2015

Show all 39 references
  1. [8]

    From few to many: Illumination cone models for face recognition under variable lighting and pose

    Athinodoros S Georghiades, Peter N Belhumeur, and David J Kriegman. From few to many: Illumination cone models for face recognition under variable lighting and pose. IEEE Transactions on Pattern Analysis and Machine Intelligence, (6):643–660, 2001. 6

    2001

  2. [9]

    On the asymptotic spectrum of random walks on infinite families of graphs

    Rostislav I Grigorchuk and Andrzej Zuk. On the asymptotic spectrum of random walks on infinite families of graphs. Random Walks and Discrete Potential Theory (Cortona, 1997), Sympos. Math, 39:188–204, 1999. 3

    1997

  3. [10]

    Wavelets on graphs via spectral graph theory

    David K Hammond, Pierre Vandergheynst, and R ´emi Gri- bonval. Wavelets on graphs via spectral graph theory. Ap- plied and Computational Harmonic Analysis , 30(2):129– 150, 2011. 2

    2011

  4. [11]

    Deep subspace clustering networks

    Pan Ji, Tong Zhang, Hongdong Li, Mathieu Salzmann, and Ian Reid. Deep subspace clustering networks. InAdvances in Neural Information Processing Systems, pages 24–33, 2017. 2, 5

    2017

  5. [12]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 ,

    arXiv

  6. [13]

    Variational graph auto- encoders

    Thomas N Kipf and Max Welling. Variational graph auto- encoders. NIPS Workshop on Bayesian Deep Learning ,

  7. [14]

    Kipf and Max Welling

    Thomas N. Kipf and Max Welling. Semi-supervised classi- fication with graph convolutional networks. In International Conference on Learning Representations, 2017. 1, 2, 3, 4

    2017

  8. [15]

    Life in the network: the coming age of computational social science

    David Lazer, Alex Sandy Pentland, Lada Adamic, Sinan Aral, Albert Laszlo Barabasi, Devon Brewer, Nicholas Christakis, Noshir Contractor, James Fowler, Myron Gut- mann, et al. Life in the network: the coming age of computational social science. Science (New York, NY) , 323(5915...

    2009

  9. [16]

    The mnist database of handwritten digits

    Yann LeCun. The mnist database of handwritten digits. http://yann. lecun. com/exdb/mnist/, 1998. 6

    1998

  10. [17]

    Learning to discover social circles in ego networks

    Jure Leskovec and Julian J Mcauley. Learning to discover social circles in ego networks. In Advances in Neural Infor- mation Processing Systems, pages 539–547, 2012. 6

    2012

  11. [19]

    Deeper insights into graph convolutional networks for semi-supervised learn- ing

    Qimai Li, Zhichao Han, and Xiao-Ming Wu. Deeper insights into graph convolutional networks for semi-supervised learn- ing. In Thirty-Second AAAI Conference on Artificial Intelli- gence, 2018. 2, 3

    2018

  12. [20]

    Deformable shape completion with graph convolu- tional autoencoders

    Or Litany, Alex Bronstein, Michael Bronstein, and Ameesh Makadia. Deformable shape completion with graph convolu- tional autoencoders. arXiv preprint arXiv:1712.00268, 2017. 1

    2017 arXiv

  13. [21]

    Least squares quantization in pcm

    Stuart Lloyd. Least squares quantization in pcm. IEEE Transactions on Information Theory , 28(2):129–137, 1982. 6, 7

    1982

  14. [22]

    Robust and efficient subspace segmentation via least squares regression

    Can-Yi Lu, Hai Min, Zhong-Qiu Zhao, Lin Zhu, De-Shuang Huang, and Shuicheng Yan. Robust and efficient subspace segmentation via least squares regression. In European con- ference on computer vision, pages 347–360. Springer, 2012. 5

    2012

  15. [23]

    Visualiz- ing data using t-sne

    Laurens van der Maaten and Geoffrey Hinton. Visualiz- ing data using t-sne. Journal of machine learning research, 9(Nov):2579–2605, 2008. 8

    2008

  16. [24]

    Ge- ometric matrix completion with recurrent multi-graph neural networks

    Federico Monti, Michael Bronstein, and Xavier Bresson. Ge- ometric matrix completion with recurrent multi-graph neural networks. In Advances in Neural Information Processing Systems, pages 3697–3707, 2017. 1

    2017

  17. [25]

    Columbia object image library (coil-20)

    Sameer A Nene, Shree K Nayar, Hiroshi Murase, et al. Columbia object image library (coil-20). 1996. 6

    1996

  18. [26]

    On spectral clustering: Analysis and an algorithm

    Andrew Y Ng, Michael I Jordan, and Yair Weiss. On spectral clustering: Analysis and an algorithm. InAdvances in Neural Information Processing Systems, pages 849–856, 2002. 1, 5, 6, 7

    2002

  19. [27]

    Adversarially regularized graph autoen- coder for graph embedding

    Shirui Pan, Ruiqi Hu, Guodong Long, Jing Jiang, Lina Yao, and Chengqi Zhang. Adversarially regularized graph autoen- coder for graph embedding. In IJCAI, pages 2609–2615,

  20. [28]

    Deep- walk: Online learning of social representations

    Bryan Perozzi, Rami Al-Rfou, and Steven Skiena. Deep- walk: Online learning of social representations. In Proceed- ings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data mining , pages 701–710. ACM, 2014. 6

  21. [29]

    Collective classifica- tion in network data

    Prithviraj Sen, Galileo Namata, Mustafa Bilgic, Lise Getoor, Brian Galligher, and Tina Eliassi-Rad. Collective classifica- tion in network data. AI Magazine, 29(3):93, 2008. 6

    2008

  22. [30]

    Normalized cuts and image segmentation

    Jianbo Shi and Jitendra Malik. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8):888–905, 2000. 1

    2000

  23. [31]

    The emerging field of sig- nal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains

    David I Shuman, Sunil K Narang, Pascal Frossard, Antonio Ortega, and Pierre Vandergheynst. The emerging field of sig- nal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Sig- nal Processing Magazine, 30(3):83–98, 2013. 2

    2013

  24. [32]

    A signal processing approach to fair surface design

    Gabriel Taubin. A signal processing approach to fair surface design. In Proceedings of the 22nd Annual Conference on Computer graphics and Interactive techniques , pages 351–

  25. [33]

    Learning deep representations for graph clustering

    Fei Tian, Bin Gao, Qing Cui, Enhong Chen, and Tie-Yan Liu. Learning deep representations for graph clustering. In AAAI, pages 1293–1299, 2014. 6

    2014

  26. [34]

    Subspace clustering

    Ren ´e Vidal. Subspace clustering. IEEE Signal Processing Magazine, 28(2):52–68, 2011. 2

    2011

  27. [35]

    Mgae: Marginalized graph autoencoder for graph clustering

    Chun Wang, Shirui Pan, Guodong Long, Xingquan Zhu, and Jing Jiang. Mgae: Marginalized graph autoencoder for graph clustering. In Proceedings of the 2017 ACM on Conference on Information and Knowledge Management , pages 889–

    2017

  28. [36]

    Robust multi- view spectral clustering via low-rank and sparse decomposi- tion

    Rongkai Xia, Yan Pan, Lei Du, and Jian Yin. Robust multi- view spectral clustering via low-rank and sparse decomposi- tion. In AAAI, pages 2149–2155, 2014. 6

    2014

  29. [37]

    Network representation learning with rich text information

    Cheng Yang, Zhiyuan Liu, Deli Zhao, Maosong Sun, and Ed- ward Y Chang. Network representation learning with rich text information. In IJCAI, pages 2111–2117, 2015. 6

    2015

  30. [38]

    Overlapping community detection at scale: a nonnegative matrix factorization ap- proach

    Jaewon Yang and Jure Leskovec. Overlapping community detection at scale: a nonnegative matrix factorization ap- proach. In Proceedings of the sixth ACM International Con- ference on Web Search and Data Mining , pages 587–596. ACM, 2013. 6

    2013

  31. [39]

    Deep adversarial subspace clustering

    Pan Zhou, Yunqing Hou, and Jiashi Feng. Deep adversarial subspace clustering. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 1596– 1604, 2018. 2

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.