Pith. sign in

REVIEW 3 major objections 6 minor 66 references

Graph Transfer Learning via Adversarial Domain Adaptation with Graph Convolution

T0 review · 3 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 AdaGCN combines graph convolution with adversarial domain adaptation to transfer node labels from a partially labeled source network to an unlabeled or partially labeled target network.

desk verdict AdaGCN is a solid empirical contribution to graph transfer learning, but its marginal-only adversarial alignment leaves class-level transfer unproven and the experiments need variance reporting and code. read the letter →

arxiv 1909.01541 v4 pith:4CQ35F3C submitted 2019-09-04 cs.LG cs.SIstat.ML

Quanyu Dai , Xiao-Ming Wu , Jiaren Xiao , Xiao Shen , Dan Wang This is my paper · ORCID
classification cs.LGcs.SIstat.ML
keywords cross-networknodeclassificationgraphtransferlearningadversarialdomainadaptationconvolutionalnetworksWassersteindistancesemi-supervisednetworkembeddinglabelefficiency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper addresses cross-network node classification: using a partially labeled source network to classify nodes in a target network that has no labels or few labels. It claims that combining graph convolutional networks, which integrate topology and attributes into node representations, with an adversarial component that minimizes the Wasserstein distance between source and target representation distributions, makes class information transferable even when the networks share few attributes. Experiments on three citation networks across six transfer tasks report consistent gains over graph convolution without adaptation and over prior network transfer baselines, with stronger gains at low source label rates. The paper also introduces an improved graph convolution layer whose smoothing strength can be tuned to ease transfer.

What carries the argument

The load-bearing mechanism is the two-player adversarial game between the GCN feature extractor (generator) and a domain critic (discriminator). The generator maps each network's adjacency matrix and node attributes to a shared representation space; the critic estimates the Wasserstein-1 distance between the source and target empirical distributions of these representations, and the generator is trained to minimize that distance while the label classifier is trained on source labels. An improved GCN layer, with an exponentiated adjacency matrix $\hat{A}^{n_I}$ as a smoothing-strength hyperparameter, strengthens Laplacian smoothing without stacking layers, which the paper argues makes representations easier to align and less prone to overfitting at low label rates.

What would settle it

A synthetic transfer task with two networks drawn from the same feature distribution but with the class labels of the target network permuted relative to the source: if AdaGCN then assigns the source label to target nodes with high confidence, the reported transfer success comes from label-conditional alignment that marginal adversarial adaptation does not guarantee.

Watch

Extended reading notes

Core claim

The central claim is that a shared-weight graph convolutional feature extractor, trained jointly with a label classifier on the source network and a Wasserstein domain critic on both networks, learns node representations that are simultaneously class-discriminative and domain-invariant, so that the source-trained classifier performs well on target nodes. The adversarial component is framed as a minimax game: the representation learner tries to fool a domain critic that estimates the Wasserstein-1 distance between source and target representation distributions, with a gradient penalty enforcing the Lipschitz constraint. With this machinery, AdaGCN and its improved variant AdaIGCN achieve state-of-the-art micro-F1 and macro-F1 in six citation-network transfer tasks in both unsupervised and semi-supervised settings, and the reported visualization shows source and target nodes of the same class clustering together after adaptation.

Load-bearing premise

Aligned marginal distributions of node representations are assumed to be enough for the source-trained classifier to work on target nodes, which requires that the aligned classes match up; if a source class aligns with a different target class, the transferred labels will be wrong.

Editorial extensions

If this is right

  • Node classification becomes possible for newly formed or sparsely labeled networks by borrowing labels from a related network with no cross-network edges.
  • Annotation cost falls: AdaGCN reports high accuracy with only 5-10% labeled source nodes, and AdaIGCN works with 5%.
  • The method tolerates low overlap between source and target attribute sets, extending transfer to networks with different feature vocabularies.
  • The framework is modular: other adversarial domain adaptation objectives could replace the Wasserstein critic without changing the overall architecture.

Reading between the lines

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

  • A natural extension is conditional adversarial alignment, conditioning the critic on predicted class, which would directly address the class-misalignment risk of marginal alignment; the paper lists this as future work.
  • The robustness claim could be tested more adversarially: deliberately swap label semantics between source and target to see whether the marginal alignment hurts rather than helps.
  • Applying AdaGCN to multiple source networks simultaneously might improve target accuracy when no single source is similar, a scenario the paper notes but does not explore.
  • The reported 26.62% jump when smoothing strength goes from 0 to 1 suggests that graph convolution itself, not just domain alignment, is doing much of the transfer work; ablating alignment while varying smoothing would isolate the two effects.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper proposes AdaGCN, a framework for cross-network node classification that combines graph convolutional networks with Wasserstein adversarial domain adaptation. The model uses two GCNs (or an improved IGCN variant) with shared parameters to learn node representations for the source and target networks, a domain critic trained with a gradient penalty to align the marginal distributions of source and target representations, and a classifier trained on source labels (with optional target labels in the semi-supervised setting). The authors evaluate on three citation networks across six transfer tasks, in both unsupervised and semi-supervised settings, and report improvements over GCN, DeepWalk, node2vec, ANRL, GraphSAGE, WDGRL, NetTr, and CDNE. They also study the effect of source/target training rates, common attribute rate, smoothing parameter, and several hyperparameters, and provide a t-SNE visualization of learned representations.

Significance. If the claims hold, this is a useful contribution to a realistic and underexplored problem: cross-network node classification without cross-network edges, with only a small amount of source labels and possibly large distribution discrepancy. The combination of graph convolution with adversarial domain adaptation is natural, and the paper provides a broad empirical comparison on real-world datasets. The authors also release code, which supports reproducibility. The main limitation is that the central empirical claim is not yet fully established: the adversarial objective aligns only marginal distributions, the reported results are single-run numbers without variance or significance tests, and the visualization covers only two of five classes on one task. These gaps are load-bearing for the paper's claim of successfully transferring class information.

major comments (3)
  1. [Tables 3 and 4] The objective minimizes the Wasserstein distance between the marginal distributions of source and target node representations, while the classifier in Eq. (6) is trained only on source labels (with optional target labels). Correct transfer requires alignment of class-conditional distributions, but marginal alignment is compatible with class permutation: source class A can be mapped to target class B without increasing the objective. The appeal to domain adaptation theory in Section 2.3 (Refs. [53], [54]) does not close this gap, because those bounds involve an ideal joint error and a hypothesis-class divergence rather than a plain marginal Wasserstein distance, and no analogous bound is derived for AdaGCN. Section 6 explicitly lists conditional adversarial domain adaptation as future work, which is an implicit acknowledgment of this limitation. The paper needs an experiment or analysis that checks class-conditional alignment, for example per-class confusion diagnostics, class-conditional MMD, or a conditional domain critic.
  2. [Tables 3 and 4] All reported numbers appear to come from a single run, with no standard deviations, confidence intervals, or significance tests. The headline gains over CDNE are modest (4.41% and 5.81% relative on average in Table 3; 3.14% and 3.55% in Table 4), so they could plausibly change with different random labeled-node splits or training seeds. The text uses the word 'significant' (e.g., Section 5.2.1) without statistical support. The authors should run multiple random trials with different labeled-node selections and report means with standard deviations, and when claiming superiority, provide a paired significance test (e.g., Wilcoxon signed-rank test across tasks or a paired t-test per task).
  3. [Section 5.7, Figure 7] The t-SNE visualization shows only two of the five classes ('Databases' and 'Computer Vision') on a single transfer task (ACMv9 to Citationv1). This is insufficient to rule out class confusion on the remaining classes or on the other five tasks. The paper should provide either per-class alignment diagnostics (e.g., class-conditional distance between source and target representations, or per-class F1 per target class) or visualizations covering all classes for representative tasks. Without such evidence, the claim that AdaGCN 'successfully transfers class information' remains only partially supported.
minor comments (6)
  1. [Abstract and Section 1] The abstract in the full text says 'Codes will be released upon acceptance,' while the arXiv abstract and the first page state that source code is already available at a GitHub URL; please make this consistent.
  2. [Section 4.2, Eq. (4)] In Eq. (4), the notation $\hat{A}^{n_I}H_g^{(k-1)}W_g^{(k)}$ could be confused with an exponent on the matrix product; please clarify that $n_I$ is the number of repeated left-multiplications of the filter on the feature matrix, as described in the text.
  3. [Section 4.5] The sentence 'the computational complexity mainly consists of three parts, including the GCN layers (Eq. (2)), the label classifier (Eq. (4)) and the domain critic (Eq. (7))' cites Eq. (4) for the label classifier, but Eq. (4) is the IGCN layer; the label classifier is defined in Eq. (5). Please correct this cross-reference.
  4. [Algorithm 1] In Algorithm 1, the notation $N_s$ and $N_t$ is used but not defined; it appears to denote numbers of sampled nodes for the critic update, but this should be stated explicitly, along with how the pairs $(h_s, h_t)$ in line 7 are sampled.
  5. [Table 2] Table 2 is not formatted correctly: the rows for Citationv1 and ACMv9 appear to have missing cells, and the intended column headers are not all visible. Please restructure the table so that #Nodes, #Edges, #Attributes, #Union Attributes, and #Labels are clearly aligned for each dataset.
  6. [Section 5.2.1] Phrases such as 'significant 13.54% and 19.03% relative gains' use 'significant' in a statistical sense without supporting tests; consider replacing with 'relative improvements' or adding the statistical analysis suggested above.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: AdaGCN's transfer performance is an empirical claim benchmarked against external baselines, and its objective does not reduce to a fitted input or to a self-citation chain.

full rationale

The paper's derivation chain is a learning objective, not an analytic derivation that re-imports its conclusion. The overall loss in Eq. (11) is min_{θg,θc} {Lc + λ max_{θd}[Ld − γLgrad]}, where Lc (Eq. 6) is the cross-entropy on labeled source nodes and Ld (Eq. 8) is the empirical Wasserstein critic loss borrowed from the external WDGRL/WGAN-GP line (Refs. [18], [60], [61]). No parameter is fitted to the target labels that are being predicted: in the headline unsupervised setting the target network is completely unlabeled, and Tables 3–4 report F1 scores measured on target nodes left out of training. The comparisons are against external and independent baselines (GCN, GraphSAGE, DeepWalk, node2vec, ANRL, DNNs, WDGRL, NetTr, CDNE). CDNE (Ref. [4]) is a co-author's prior work and IGCN (Ref. [19]) is a co-author's architectural proposal, but both are used as starting points or baselines rather than as authority that forces the paper's result; the paper's advantage over CDNE and the benefit of nI smoothing are established by its own experiments, not imported from the citations. The domain-adaptation theory cited in Section 4.4 (Refs. [53], [54]) motivates the adversarial objective but is not invoked as a uniqueness theorem, and no alternative method is ruled out by a self-citation. The known weakness that marginal Wasserstein alignment does not guarantee class-conditional alignment is a correctness and generalization risk, not circularity: an incorrect class-permuting alignment would be a failure of the method, not an equivalence between the method's output and its input. Section 6 listing conditional alignment as future work confirms the limitation but does not make the reported transfer a renamed fit or a self-referential prediction.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The framework rests on standard domain adaptation theory, a shared label space assumption, and a smoothing-based representation learning choice. The free parameters are hyperparameters tuned on the target tasks, not fitted constants in a derivation. No new physical or mathematical entities are introduced.

free parameters (5)
  • Smoothing parameter nI = 10
    Set to 10 for all tasks in AdaIGCN to control Laplacian smoothing strength in the improved GCN layer. Chosen based on sensitivity analysis in Section 5.5 and prior work [19].
  • Domain adaptation coefficient lambda = 1
    Balances classification loss and domain adaptation loss in Eq. (11). Tuned in Section 5.6; performance drops when lambda deviates from 1.
  • Gradient penalty coefficient gamma = 10
    Weight for the Lipschitz gradient penalty in Eq. (10). Set to 10 following [61] and confirmed in Section 5.6.
  • Domain critic training step nd = 10
    Number of critic updates per generator update in Algorithm 1. Increasing nd from 5 to 10 improves performance, then stabilizes (Section 5.6).
  • Learning rates alpha1, alpha2 = 1.5e-3
    Used with Adam for both the critic and the generator/classifier; fixed for all experiments (Section 5.1.3).
assumptions (6)
  • domain assumption The source and target networks share the same label space Y and the same underlying label function after domain alignment
    Section 3 defines the problem with the same label set and assumes a classifier trained on source representations can classify target representations once distributions are aligned. This is not proven in the paper.
  • domain assumption Adversarial alignment of marginal node representation distributions reduces target classification error
    Section 4.4 relies on domain adaptation theory from [53], [54]. The theory bounds target error by source error plus distribution divergence plus a term that is assumed small; the paper does not verify that term.
  • domain assumption Marginal alignment without class-conditional alignment is sufficient for successful transfer
    The Wasserstein loss in Eq. (7) aligns only the marginal distributions Phs and Pht. The paper does not use target labels or label information in the critic, so class-conditional alignment is not enforced; this assumption is load-bearing for the unsupervised setting.
  • domain assumption Low-pass graph filtering produces representations that ease knowledge transfer
    Section 4.2 states that with a proper low-pass filter, graph convolution generates useful representations for transfer, citing [19]. This is a heuristic rather than a proven property for cross-network transfer.
  • standard math The empirical Wasserstein distance with gradient penalty is a valid approximation of W1
    Section 4.4 uses the Kantorovich-Rubinstein duality and the gradient penalty from [61], which is standard in WGAN literature.
  • domain assumption Nodes can be treated as i.i.d. samples by the domain critic despite graph correlations
    The critic loss in Eq. (8) sums over individual node representations. The paper itself notes in Section 2.3 that graph entities are correlated and violate the i.i.d. assumption, so this is an acknowledged approximation without dedicated analysis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph Transfer Learning via Adversarial Domain Adaptation with Graph Convolution." pith.science (2026). https://pith.science/paper/4CQ35F3C

@misc{pith2026190901541,
  author       = {Pith},
  title        = {Pith review of: Graph Transfer Learning via Adversarial Domain Adaptation with Graph Convolution},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4CQ35F3C}},
  note         = {Machine review of arXiv:1909.01541}
}
read the original abstract

This paper studies the problem of cross-network node classification to overcome the insufficiency of labeled data in a single network. It aims to leverage the label information in a partially labeled source network to assist node classification in a completely unlabeled or partially labeled target network. Existing methods for single network learning cannot solve this problem due to the domain shift across networks. Some multi-network learning methods heavily rely on the existence of cross-network connections, thus are inapplicable for this problem. To tackle this problem, we propose a novel \textcolor{black}{graph} transfer learning framework AdaGCN by leveraging the techniques of adversarial domain adaptation and graph convolution. It consists of two components: a semi-supervised learning component and an adversarial domain adaptation component. The former aims to learn class discriminative node representations with given label information of the source and target networks, while the latter contributes to mitigating the distribution divergence between the source and target domains to facilitate knowledge transfer. Extensive empirical evaluations on real-world datasets show that AdaGCN can successfully transfer class information with a low label rate on the source network and a substantial divergence between the source and target domains. The source code for reproducing the experimental results is available at https://github.com/daiquanyu/AdaGCN.

Figures

Figures reproduced from arXiv: 1909.01541 by the authors.

Figure 1
Figure 1. Cross-network node classification. We aim to transfer knowledge [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Model architecture of AdaGCN. Specifically, there exists a domain divergence between the source and target networks, i.e., Ds 6= Dt , but the label space Y = {1, · · · , L} is the same, and our goal is to learn a classifier f to accurately classify the nodes in the target network with the assistance of the partially labeled source network. 4 PROPOSED METHOD 4.1 An Overview of Model Architecture To solve cross-networ… view at source ↗
Figure 3
Figure 3. Multi-label classification with varying source training rates. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Multi-label classification with varying target training rates. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Multi-label classification on Citationv1 with varying common [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 7
Figure 7. Figure 7: Visualization of the learned node representations from [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

66 extracted references · 58 canonical work pages

  1. [53]

    Analysis of representations for domain adaptation,

    S. Ben-David, J. Blitzer, K. Crammer, and F. Pereira, “Analysis of representations for domain adaptation,” inNIPS. MIT Press, 2006, pp. 137–144

    work page 2006

  2. [54]

    A theory of learning from different domains,

    S. Ben-David, J. Blitzer, K. Crammer, A. Kulesza, F. Pereira, and J. W. Vaughan, “A theory of learning from different domains,” Machine Learning, vol. 79, no. 1-2, pp. 151–175, 2010

    work page 2010

  3. [1]

    Node classifica- tion in social networks,

    S. Bhagat, G. Cormode, and S. Muthukrishnan, “Node classifica- tion in social networks,” in Social Network Data Analytics, 2011, pp. 115–148

    work page 2011

  4. [2]

    Small sample learning in big data era,

    J. Shu, Z. Xu, and D. Meng, “Small sample learning in big data era,” CoRR, vol. abs/1808.04572, 2018

    arXiv 2018

  5. [3]

    Transfer learning across networks for collective classification,

    M. Fang, J. Yin, and X. Zhu, “Transfer learning across networks for collective classification,” in ICDM, 2013, pp. 161–170

    work page 2013

  6. [4]

    Network embedding for cross-network node classification,

    X. Shen and F. Chung, “Network embedding for cross-network node classification,” in arXiv:1901.07264, 2019

    arXiv 1901

  7. [5]

    Deepwalk: online learning of social representations,

    B. Perozzi, R. Al-Rfou, and S. Skiena, “Deepwalk: online learning of social representations,” in KDD, 2014, pp. 701–710

    work page 2014

  8. [6]

    LINE: large-scale information network embedding,

    J. Tang, M. Qu, M. Wang, M. Zhang, J. Yan, and Q. Mei, “LINE: large-scale information network embedding,” in WWW, 2015, pp. 1067–1077

    work page 2015

Show all 66 references
  1. [7]

    node2vec: Scalable feature learning for networks,

    A. Grover and J. Leskovec, “node2vec: Scalable feature learning for networks,” in KDD, 2016, pp. 855–864

    2016

  2. [8]

    On generalizing neural node em- bedding methods to multi-network problems,

    M. Heimann and D. Koutra, “On generalizing neural node em- bedding methods to multi-network problems,” in KDD MLG Workshop, 2017

    2017

  3. [9]

    Revisiting semi- supervised learning with graph embeddings,

    Z. Yang, W. W. Cohen, and R. Salakhutdinov, “Revisiting semi- supervised learning with graph embeddings,” in ICML, 2016, pp. 40–48

    2016

  4. [10]

    Semi-supervised classification with graph convolutional networks,

    T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in ICLR, 2017

    2017

  5. [11]

    Inductive representa- tion learning on large graphs,

    W. L. Hamilton, R. Ying, and J. Leskovec, “Inductive representa- tion learning on large graphs,” in NIPS, 2017

    2017

  6. [12]

    Graph attention networks,

    P . Velickovic, G. Cucurull, A. Casanova, A. Romero, P . Li `o, and Y. Bengio, “Graph attention networks,” CoRR, 2017

    2017

  7. [13]

    Co-regularized deep multi-network embedding,

    J. Ni, S. Chang, X. Liu, W. Cheng, H. Chen, D. Xu, and X. Zhang, “Co-regularized deep multi-network embedding,” in WWW, 2018, pp. 469–478. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, SUBMISSION 2019 13

    2018

  8. [14]

    Embedding of embedding (EOE): joint embedding for coupled heterogeneous networks,

    L. Xu, X. Wei, J. Cao, and P . S. Yu, “Embedding of embedding (EOE): joint embedding for coupled heterogeneous networks,” in WSDM, 2017, pp. 741–749

    2017

  9. [15]

    A survey on transfer learning,

    S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Trans. Knowl. Data Eng., vol. 22, no. 10, pp. 1345–1359, 2010

    2010

  10. [16]

    Deep visual domain adaptation: A survey,

    M. Wang and W. Deng, “Deep visual domain adaptation: A survey,” Neurocomputing, vol. 312, pp. 135–153, 2018

    2018

  11. [17]

    Domain- adversarial training of neural networks,

    Y. Ganin, E. Ustinova, H. Ajakan, P . Germain, H. Larochelle, F. Laviolette, M. Marchand, and V . S. Lempitsky, “Domain- adversarial training of neural networks,” JMLR, vol. 17, pp. 59:1– 59:35, 2016

    2016

  12. [18]

    Wasserstein distance guided representation learning for domain adaptation,

    J. Shen, Y. Qu, W. Zhang, and Y. Yu, “Wasserstein distance guided representation learning for domain adaptation,” in AAAI, 2018, pp. 4058–4065

    2018

  13. [19]

    Label efficient semi- supervised learning via graph filtering,

    Q. Li, X. Wu, H. Liu, X. Zhang, and Z. Guan, “Label efficient semi- supervised learning via graph filtering,” in CVPR, 2019

    2019

  14. [20]

    Generative adversarial nets,

    I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde- Farley, S. Ozair, A. C. Courville, and Y. Bengio, “Generative adversarial nets,” in NIPS, 2014, pp. 2672–2680

    2014

  15. [21]

    Grarep: Learning graph representations with global structural information,

    S. Cao, W. Lu, and Q. Xu, “Grarep: Learning graph representations with global structural information,” in CIKM, 2015, pp. 891–900

    2015

  16. [22]

    Community preserving network embedding,

    X. Wang, P . Cui, J. Wang, J. Pei, W. Zhu, and S. Yang, “Community preserving network embedding,” in AAAI, 2017, pp. 203–209

    2017

  17. [23]

    Deep neural networks for learning graph representations,

    S. Cao, W. Lu, and Q. Xu, “Deep neural networks for learning graph representations,” in AAAI, 2016, pp. 1145–1152

    2016

  18. [24]

    Structural deep network embed- ding,

    D. Wang, P . Cui, and W. Zhu, “Structural deep network embed- ding,” in KDD, 2016, pp. 1225–1234

    2016

  19. [25]

    Deep network embedding with aggregated proximity preserving,

    X. Shen and F. Chung, “Deep network embedding with aggregated proximity preserving,” in ASONAM, 2017, pp. 40–43

    2017

  20. [26]

    Deep network embedding for graph representation learn- ing in signed networks,

    ——, “Deep network embedding for graph representation learn- ing in signed networks,” IEEE Transactions on Cybernetics, 2018

    2018

  21. [27]

    Adversarial network embed- ding,

    Q. Dai, Q. Li, J. Tang, and D. Wang, “Adversarial network embed- ding,” in AAAI, 2018

    2018

  22. [28]

    Adver- sarially regularized graph autoencoder for graph embedding,

    S. Pan, R. Hu, G. Long, J. Jiang, L. Yao, and C. Zhang, “Adver- sarially regularized graph autoencoder for graph embedding,” in IJCAI, 2018

    2018

  23. [29]

    Adversarial training methods for network embedding,

    Q. Dai, X. Shen, L. Zhang, Q. Li, and D. Wang, “Adversarial training methods for network embedding,” in WWW, 2019, pp. 329–339

    2019

  24. [30]

    ANRL: attributed network representation learning via deep neural networks,

    Z. Zhang, H. Yang, J. Bu, S. Zhou, P . Yu, J. Zhang, M. Ester, and C. Wang, “ANRL: attributed network representation learning via deep neural networks,” in IJCAI, 2018, pp. 3155–3161

    2018

  25. [31]

    On exploring semantic meanings of links for embedding social networks,

    L. Xu, X. Wei, J. Cao, and P . S. Yu, “On exploring semantic meanings of links for embedding social networks,” inWWW, 2018, pp. 479–488

    2018

  26. [32]

    Max-margin deepwalk: Discriminative learning of network representation,

    C. Tu, W. Zhang, Z. Liu, and M. Sun, “Max-margin deepwalk: Discriminative learning of network representation,” inIJCAI, 2016

    2016

  27. [33]

    Tri-party deep network representation,

    S. Pan, J. Wu, X. Zhu, C. Zhang, and Y. Wang, “Tri-party deep network representation,” in IJCAI, 2016, pp. 1895–1901

    2016

  28. [34]

    Label informed attributed network embedding,

    X. Huang, J. Li, and X. Hu, “Label informed attributed network embedding,” in WSDM, 2017, pp. 731–739

    2017

  29. [35]

    Semi-supervised embedding in attributed networks with outliers,

    J. Liang, P . Jacobs, J. Sun, and S. Parthasarathy, “Semi-supervised embedding in attributed networks with outliers,” in SDM, 2018

    2018

  30. [36]

    Aligning users across social networks using network embedding,

    L. Liu, W. K. Cheung, X. Li, and L. Liao, “Aligning users across social networks using network embedding,” in IJCAI, 2016, pp. 1774–1780

    2016

  31. [37]

    REGAL: rep- resentation learning-based graph alignment,

    M. Heimann, H. Shen, T. Safavi, and D. Koutra, “REGAL: rep- resentation learning-based graph alignment,” in CIKM, 2018, pp. 117–126

    2018

  32. [38]

    Inferring social ties across heterogenous networks,

    J. Tang, T. Lou, and J. M. Kleinberg, “Inferring social ties across heterogenous networks,” in WSDM, 2012, pp. 743–752

    2012

  33. [39]

    Leveraging cross-network in- formation for graph sparsification in influence maximization,

    X. Shen, F. Chung, and S. Mao, “Leveraging cross-network in- formation for graph sparsification in influence maximization,” in SIGIR, 2017, pp. 801–804

    2017

  34. [40]

    Cross-network learning with fuzzy labels for seed selection and graph sparsification in influence maximization,

    ——, “Cross-network learning with fuzzy labels for seed selection and graph sparsification in influence maximization,” IEEE Trans- actions on Fuzzy Systems, 2019

    2019

  35. [41]

    Predicting positive and negative links in signed social networks by transfer learning,

    J. Ye, H. Cheng, Z. Zhu, and M. Chen, “Predicting positive and negative links in signed social networks by transfer learning,” in WWW, 2013

    2013

  36. [42]

    Transfer learning for deep learning on graph-structured data,

    J. Lee, H. Kim, J. Lee, and S. Yoon, “Transfer learning for deep learning on graph-structured data,” in AAAI, 2017, pp. 2154–2160

    2017

  37. [43]

    Distant domain transfer learning,

    B. Tan, Y. Zhang, S. J. Pan, and Q. Yang, “Distant domain transfer learning,” in AAAI, 2017, pp. 2604–2610

    2017

  38. [44]

    Beyond sharing weights for deep domain adaptation,

    A. Rozantsev, M. Salzmann, and P . Fua, “Beyond sharing weights for deep domain adaptation,” CoRR, vol. abs/1603.06432, 2016

    2016 arXiv

  39. [45]

    Learning transferable features with deep adaptation networks,

    M. Long, Y. Cao, J. Wang, and M. I. Jordan, “Learning transferable features with deep adaptation networks,” in ICML, 2015, pp. 97– 105

    2015

  40. [46]

    Deep CORAL: correlation alignment for deep domain adaptation,

    B. Sun and K. Saenko, “Deep CORAL: correlation alignment for deep domain adaptation,” in ECCV, 2016, pp. 443–450

    2016

  41. [47]

    Deep domain confusion: Maximizing for domain invariance,

    E. Tzeng, J. Hoffman, N. Zhang, K. Saenko, and T. Darrell, “Deep domain confusion: Maximizing for domain invariance,” CoRR, vol. abs/1412.3474, 2014

    2014 arXiv

  42. [48]

    Supervised rep- resentation learning: Transfer learning with deep autoencoders,

    F. Zhuang, X. Cheng, P . Luo, S. J. Pan, and Q. He, “Supervised rep- resentation learning: Transfer learning with deep autoencoders,” in IJCAI, 2015, pp. 4119–4125

    2015

  43. [49]

    Learning to discover cross-domain relations with generative adversarial net- works,

    T. Kim, M. Cha, H. Kim, J. K. Lee, and J. Kim, “Learning to discover cross-domain relations with generative adversarial net- works,” in ICML, 2017, pp. 1857–1865

    2017

  44. [50]

    Adversarial discriminative domain adaptation,

    E. Tzeng, J. Hoffman, K. Saenko, and T. Darrell, “Adversarial discriminative domain adaptation,” in CVPR, 2017, pp. 2962–2971

    2017

  45. [51]

    Multi-adversarial domain adaptation,

    Z. Pei, Z. Cao, M. Long, and J. Wang, “Multi-adversarial domain adaptation,” in AAAI, 2018, pp. 3934–3941

    2018

  46. [52]

    Conditional adver- sarial domain adaptation,

    M. Long, Z. Cao, J. Wang, and M. I. Jordan, “Conditional adver- sarial domain adaptation,” in NeurIPS, 2018, pp. 1647–1657

    2018

  47. [55]

    Attributed graph clustering via adaptive graph convolution,

    X. Zhang, H. Liu, Q. Li, and X. Wu, “Attributed graph clustering via adaptive graph convolution,” in Proceedings of the Twenty- Eighth International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China, August 10-16, 2019 , 2019, pp. 4327–4333

    2019

  48. [56]

    Graph neural networks for social recommendation,

    W. Fan, Y. Ma, Q. Li, Y. He, Y. E. Zhao, J. Tang, and D. Yin, “Graph neural networks for social recommendation,” in WWW, 2019, pp. 417–426

    2019

  49. [57]

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

    D. I. Shuman, S. K. Narang, P . Frossard, A. Ortega, and P . Van- dergheynst, “The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains,” IEEE Signal Processing Magazine, vol. 30, no. 3, pp. 83–98, 2013

    2013

  50. [58]

    Discrete signal processing on graphs,

    A. Sandryhaila and J. M. F. Moura, “Discrete signal processing on graphs,” IEEE Trans. Signal Processing, vol. 61, no. 7, pp. 1644–1656, 2013

    2013

  51. [59]

    Deeper insights into graph convolu- tional networks for semi-supervised learning,

    Q. Li, Z. Han, and X. Wu, “Deeper insights into graph convolu- tional networks for semi-supervised learning,” in AAAI, 2018, pp. 3538–3545

    2018

  52. [60]

    Wasserstein generative adversarial networks,

    M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein generative adversarial networks,” in ICML, 2017, pp. 214–223

    2017

  53. [61]

    Improved training of wasserstein gans,

    I. Gulrajani, F. Ahmed, M. Arjovsky, V . Dumoulin, and A. C. Courville, “Improved training of wasserstein gans,” in NIPS, 2017

    2017

  54. [62]

    Villani, Optimal transport: old and new

    C. Villani, Optimal transport: old and new . Springer-Verlag Berlin Heidelberg, 2008, vol. 338

    2008

  55. [63]

    Stochastic training of graph convo- lutional networks with variance reduction,

    J. Chen, J. Zhu, and L. Song, “Stochastic training of graph convo- lutional networks with variance reduction,” in ICML, vol. 80, 2018, pp. 941–949

    2018

  56. [64]

    Fastgcn: Fast learning with graph convolutional networks via importance sampling,

    J. Chen, T. Ma, and C. Xiao, “Fastgcn: Fast learning with graph convolutional networks via importance sampling,” in ICLR, 2018

    2018

  57. [65]

    Arnetminer: extraction and mining of academic social networks,

    J. Tang, J. Zhang, L. Yao, J. Li, L. Zhang, and Z. Su, “Arnetminer: extraction and mining of academic social networks,” inKDD, 2008, pp. 990–998

    2008

  58. [66]

    Visualizing data using t-sne,

    L. van der Maaten and G. Hinton, “Visualizing data using t-sne,” JMLR, vol. 9, pp. 2579–2605, 2008. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, SUBMISSION 2019 14 Quanyu Dai received the B.Eng. degree from the Department of Electronic Engineering, Shanghai Jiao Tong U...

    2008

Pith tools

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