Pith. sign in

REVIEW 3 major objections 6 minor 64 references

Federated Contrastive Learning of Graph-Level Representations

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a two-level contrastive framework, FCLG, can learn unsupervised graph-level representations in federated settings and reports clustering gains of 2–7% over federated baselines.

desk verdict FCLG is a reasonable first pass at unsupervised federated graph-level representation learning, but the claimed 2-7% gains currently rest on a tuning protocol that selects hyperparameters on test labels, with the appendix and code missing. read the letter →

arxiv 2411.12098 v1 pith:KZRBHW5G submitted 2024-11-18 cs.LG

Xiang Li , Gagan Agrawal , Rajiv Ramnath , Ruoming Jin This is my paper · ORCID
classification cs.LG
keywords federatedlearninggraph-levelrepresentationcontrastiveunsupervisedgraphclusteringnon-IIDdataneuralnetworksknowledgedistillation
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

Graph-level representations—embeddings of entire graphs rather than individual nodes—are needed for tasks like predicting protein properties or flagging malicious network traffic, but the graphs often cannot be pooled into one place. This paper tries to establish that such representations can still be learned without labels when data are scattered across isolated clients, by applying contrastive learning at two levels. Inside each client, augmented views of the same graph are pulled together and other graphs are pushed apart; between client and server, the current local representation is pulled toward the global model and away from the local model's previous round, which the paper argues counters non-IID drift. On four public protein and molecule benchmarks, FCLG reports 2–7% clustering accuracy gains over federated baselines in non-IID splits and 2–4% gains in IID splits. If this holds, institutions with private graph data could jointly train reusable graph embeddings without sharing raw graphs.

What carries the argument

The load-bearing mechanism is the two-level contrastive loss. Intra-contrasting generates a second view of each input graph through graph diffusion, treats that view as the sole positive pair, and uses all other graphs in the batch as negatives, following the SimCLR objective, which forces the local encoder to produce well-separated graph embeddings. Inter-contrasting compares three graph-level representations of the same batch: the current local model's output, the local model's output from the previous local epoch, and the global model's output, minimizing a contrastive loss that increases similarity between current-local and global while decreasing similarity between current-local and previous-local. This second term is the paper's answer to non-IID drift: it constrains each client's local updates to stay aligned with the aggregated global view while preventing the local model from wandering along its own skewed distribution.

What would settle it

Rerun the non-IID comparison with a single fixed hyperparameter configuration chosen without access to test labels—either identical across datasets or selected by an unsupervised criterion—and check whether FCLG's 2–7% accuracy margins over InfoGraph-FedProx survive; if they shrink to within run-to-run variance, the superiority claim fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that a single two-level contrastive objective solves both the unsupervised-learning problem and the federated non-IID problem at once. The first level, intra-contrasting, runs instance-wise contrastive discrimination on graph-level representations inside each client, so each graph becomes distinguishable from all other graphs in the local batch. The second level, inter-contrasting, uses a cosine-similarity contrastive loss that pulls the current local representation toward the global model's representation and pushes it away from the representation produced by the local model in the previous local epoch. The paper argues this inter term is an advanced ensembling-distillation technique and shows that removing it costs about 5% on DHFR and 2% on ENZYMES, while replacing it with KL or MSE distillation costs 3–5% on PROTEINS and DHFR. Across four datasets, FCLG and its node-level variant FCLG-H outperform InfoGraph and MVGRL combined with FedAvg or FedProx, with the largest margins in non-IID settings.

Load-bearing premise

The reported margins rest on the assumption that tuning all methods' hyperparameters on each dataset's true labels is a fair comparison, and that four small protein and molecule benchmarks with class-proportion skew capture the federated graph-level setting.

Editorial extensions

If this is right

  • Private graph holders can jointly train an unsupervised graph encoder without sharing raw graphs, and the learned embeddings separate classes well enough to beat locally trained federated baselines in K-Means clustering.
  • The inter-contrastive term is doing real work: dropping it lowers accuracy by about 5% on DHFR and 2% on ENZYMES, so the margin is not solely from the intra-client contrastive loss.
  • The method's advantage grows as client distributions become more skewed, since the paper reports larger relative gains at higher EMD values.
  • Using node-level rather than graph-level representations for inter-contrasting gives similar clustering performance, so the benefit does not depend on the pooling granularity of the contrastive alignment.

Reading between the lines

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

  • Our inference: the same two-level recipe should transfer to other decentralized unsupervised settings—time series, point clouds, or text—because only the graph-diffusion augmentation is domain-specific, while both contrastive losses are domain-agnostic.
  • Our inference: the knowledge-distillation framing suggests a cheaper or stronger variant might replace the previous-local-round negative with a server-side prototype or a moving average of global representations; comparing those variants would isolate what the temporal negative contributes.
  • Our inference: because the four benchmarks are small and skewed by class label only, the open question is whether the margins survive on large federated graph collections where client drift is driven by topology rather than label proportions.
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 / 6 minor

Summary. This paper proposes FCLG, a federated learning framework for unsupervised graph-level representation learning. FCLG combines two contrastive objectives: an intra-client instance-wise contrastive loss over augmented views of graphs (adapted from SimCLR/MVGRL-style augmentation) and an inter-client model-level contrastive loss that aligns the current local model's graph-level representations with the global model's representations while repelling the previous local epoch's representations (adapted from MOON). The authors evaluate the learned embeddings via K-Means clustering on four TUDatasets (PROTEINS, ENZYMES, DHFR, NCI1) in both IID and non-IID federated settings, comparing against InfoGraph and MVGRL combined with FedAvg and FedProx. They report 2-7% clustering accuracy gains over the closest baselines in non-IID settings and 2-4% in IID settings, alongside ablations and robustness experiments.

Significance. The contribution is potentially significant: if the empirical claims hold, this is an early demonstration that unsupervised graph-level representation learning can be federated without labels, and the two-level contrastive design is a sensible combination of existing ideas. The paper's strengths include the clear algorithmic description, the use of 10-run repeats for main tables, and the ablation of inter-contrast variants (Intra-FedAvg, Intra-KL, Intra-MSE). However, the empirical validation has a load-bearing flaw: hyperparameters are tuned to maximize the evaluation metric on the same dataset labels used for the reported results, and the detailed settings are relegated to a missing appendix. The 2-7% margins can therefore not currently be attributed to the method rather than to the tuning protocol. The absence of significance tests and error bars further weakens the 'significant gain' claim.

major comments (3)
  1. [IV-A and V-A] The central empirical claim is undermined by the hyperparameter selection protocol. Section IV-A states that 'all hyper-parameters are tuned for optimal clustering performance on each dataset,' and the caption of Figure 4a/4b says each result is 'obtained by fine-tuning hyper-parameters separately.' Because the metric used for tuning (clustering accuracy and macro F1 against ground-truth labels) is the same as the headline evaluation metric, the reported gains in Tables II and III (e.g., 7% on PROTEINS, ~5% on ENZYMES and DHFR, 2% on NCI1) are upper envelopes over the hyperparameter grid rather than the performance of a defensible fixed configuration. To support the claim of 'significant clustering gain,' the authors should either fix hyperparameters in advance, select them on a validation set without using the evaluation labels, or show that the reported margins are stable across the tuning grid.
  2. [IV-A] The manuscript is not reproducible as submitted. The text refers twice to an Appendix for 'detailed hyper-parameter settings' and for the official baseline code, but no appendix is included in the manuscript. This is particularly problematic because FCLG introduces additional hyperparameters (τ, τ′, the number of local epochs E, and the loss combination in Eq. 9) beyond the baselines. Without the actual grid, the reader cannot assess whether FCLG and the baselines received comparable tuning budgets, which is exactly the point on which the headline comparison rests. The authors should include the appendix and release code (or at least the full configuration and search ranges) before the results can be verified.
  3. [IV-B] The claim of a 'significant clustering gain' is not backed by statistical inference. The 10-run ranges in Tables II and III show that the mean differences are large relative to the spread, but no significance test is reported, and the ranges are conditional on the per-dataset hyperparameter selection discussed above. In addition, Figures 2, 4, and 5 present trajectories and comparisons without any error bands, so the robustness of the skew (Fig. 4a) and client-number (Fig. 4b) experiments cannot be assessed. The authors should add significance tests and error bars/bands, or at least report per-run values for the experiments in Figures 4 and 5.
minor comments (6)
  1. [Table III] The row label 'FGCL' should be 'FCLG' to match the method name used throughout the paper.
  2. [Figure 5 caption] The caption says 'on PROTEINS (20 communication rounds)' but the figure contains four lines labeled PROTEINS, ENZYMES, DHFR, and NCI1; the caption should either list all datasets or state that all four are shown.
  3. [Section IV-B] The text describing FCLG-H says 'intra-contrasting is conducted between node-level representations H via Eq. 8,' but Eq. 8 defines an inter-contrasting variant (l^H_inter); the text should say 'inter-contrasting' rather than 'intra-contrasting.'
  4. [Eq. (5)] The contrastive loss in Eq. (5) is written as a positive log ratio; to align with the standard InfoNCE formulation and the statement that the objective is minimized, it would be clearer to write it with an explicit negative sign or to state explicitly that minimizing L maximizes agreement between positive pairs.
  5. [Section IV-A] The procedure for creating the non-IID client splits is not described in the main text (e.g., whether a Dirichlet allocation over classes is used); since the non-IID setting is central to the evaluation, this detail should be provided in the main text or in the appendix.
  6. [Section IV-B] The clustering evaluation protocol does not state how the K-Means cluster labels are matched to ground-truth classes (e.g., Hungarian matching); this step should be described for reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FCLG's losses are adopted from external cited work and its central claims are evaluated on external TUDataset benchmarks; the tuning-on-test-labels caveat is an evaluation robustness concern, not a circular reduction.

full rationale

FCLG does not contain a derivation chain in which outputs are equivalent to inputs. The intra-contrastive loss (Eqs. 5-6) instantiates the SimCLR instance-wise contrastive objective from Chen et al. [30] at the graph level, using graph-diffusion augmentation following MVGRL [29]; the inter-contrastive loss (Eq. 7) is the MOON model-contrastive loss from Li et al. [32] adapted to graph-level embeddings. Neither equation defines the model in terms of downstream clustering labels, and no parameter is fitted directly to the reported accuracy or macro-F1 values. The claimed 2-7% gains are empirical comparisons on four external TUDatasets against baselines constructed from InfoGraph/MVGRL combined with FedAvg/FedProx, so the central claim is externally falsifiable rather than forced by construction. The cited MOON paper has no author overlap with the present paper, and there is no load-bearing self-citation chain. The only substantive caveats are the statements in Section IV-A that 'all hyper-parameters tuned for optimal clustering performance on each dataset' and in Section V-A that each Figure 4 result is 'obtained by fine-tuning hyper-parameters separately,' together with the reference to a hyperparameter appendix that is not present in the manuscript. Tuning hyperparameters on evaluation labels can overstate absolute performance and is a legitimate protocol weakness, but it does not make FCLG's reported gain equivalent to its inputs: the same protocol is applied to the baselines, and the model never uses labels during training. These concerns belong to evaluation robustness and reproducibility, not to circularity of the method's derivation.

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

The central claim does not rest on a derivation, so there are no free parameters in a mathematical model. The parameters that matter are training hyperparameters, all tuned for test clustering performance, which is a mild form of selection. No invented entities are introduced; all losses and architectures are prior work. The axioms are empirical assumptions about augmentation, representation quality, aggregation, and evaluation.

free parameters (5)
  • Intra-contrast temperature tau = not reported in main text
    Appears in Eq.5; controls sharpness of intra-contrast loss; tuned per dataset per Section IV-A.
  • Inter-contrast temperature tau-prime = not reported in main text
    Appears in Eq.7; controls inter-contrast loss; tuned per dataset.
  • Local training epochs E and communication rounds T = not reported in main text
    Inputs to Algorithm 1; Figure 5 shows clustering accuracy varies strongly with E, so these settings affect the central results.
  • GIN architecture and optimizer settings = not reported in main text
    Section III-D chooses a GIN encoder with L layers, sum pooling, and AdamW; exact depth, hidden size, learning rate, and weight decay are deferred to the appendix.
  • Per-dataset hyperparameter configuration = dataset-specific
    Section IV-A states all hyperparameters are tuned for optimal clustering performance on each dataset, which couples the configuration to the evaluation data.
assumptions (5)
  • domain assumption Graph diffusion augmentation preserves the semantic class of each graph, so augmented views are valid positives for contrastive learning.
    Section III.C creates augmented views via graph diffusion following [29]. If diffusion alters class-relevant structure, intra-contrast may discard useful signal.
  • domain assumption Minimizing the sum of intra- and inter-contrast losses yields representations useful for downstream clustering.
    This is the core empirical hypothesis of Section III.D; no convergence or representation guarantee is proven.
  • domain assumption Model parameter averaging as in FedAvg (Eq.1) is a meaningful aggregation for heterogeneous GNN clients.
    Used in Algorithm 1; FedAvg is known to be suboptimal under non-IID data, but is treated as sufficient here.
  • domain assumption Class-skew measured by EMD is an adequate model of non-IID data distribution across graph clients.
    Section IV.A defines EMD over graph-level class distributions; other distribution shifts such as feature or structural shift are not considered.
  • domain assumption Ground-truth class labels in TUDatasets are valid targets for evaluating unsupervised clustering.
    Evaluation uses accuracy and F1 against dataset labels; this assumes the label structure is what the embeddings should recover.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Federated Contrastive Learning of Graph-Level Representations." pith.science (2026). https://pith.science/paper/KZRBHW5G

@misc{pith2026241112098,
  author       = {Pith},
  title        = {Pith review of: Federated Contrastive Learning of Graph-Level Representations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KZRBHW5G}},
  note         = {Machine review of arXiv:2411.12098}
}
read the original abstract

Graph-level representations (and clustering/classification based on these representations) are required in a variety of applications. Examples include identifying malicious network traffic, prediction of protein properties, and many others. Often, data has to stay in isolated local systems (i.e., cannot be centrally shared for analysis) due to a variety of considerations like privacy concerns, lack of trust between the parties, regulations, or simply because the data is too large to be shared sufficiently quickly. This points to the need for federated learning for graph-level representations, a topic that has not been explored much, especially in an unsupervised setting. Addressing this problem, this paper presents a new framework we refer to as Federated Contrastive Learning of Graph-level Representations (FCLG). As the name suggests, our approach builds on contrastive learning. However, what is unique is that we apply contrastive learning at two levels. The first application is for local unsupervised learning of graph representations. The second level is to address the challenge associated with data distribution variation (i.e. the ``Non-IID issue") when combining local models. Through extensive experiments on the downstream task of graph-level clustering, we demonstrate FCLG outperforms baselines (which apply existing federated methods on existing graph-level clustering methods) with significant margins.

Figures

Figures reproduced from arXiv: 2411.12098 by the authors.

Figure 1
Figure 1. Schematic of FCLG framework architecture. Given [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Accuracy Vs. Communication Round on PROTEINS [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. T-SNE for graph-level embeddings of PROTEINS at different communication round during FCLG training. Each color [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Accuracy Vs. varying data distribution on PROTEINS, [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: FCLG Accuracy Vs. No. of Local Train Epoch on [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 47 canonical work pages

  1. [1]

    Survey of graph database models,

    R. Angles and C. Gutierrez, “Survey of graph database models,” ACM Comput. Surv. , vol. 40, no. 1, feb 2008. [Online]. Available: https://doi.org/10.1145/1322432.1322433

    arXiv 2008

  2. [2]

    Cgnn: Traffic classification with graph neural network,

    B. Pang, Y . Fu, S. Ren, Y . Wang, Q. Liao, and Y . Jia, “Cgnn: Traffic classification with graph neural network,” 2021

    work page 2021

  3. [3]

    Traffic classification based on graph convolutional network,

    X. Ji and Q. Meng, “Traffic classification based on graph convolutional network,” in 2020 IEEE International Conference on Advances in Electrical Engineering and Computer Applications( AEECA) , 2020, pp. 596–601

    work page 2020

  4. [4]

    Learned protein embeddings for machine learning,

    K. K. Yang, Z. Wu, C. N. Bedbrook, and F. H. Arnold, “Learned protein embeddings for machine learning,” Bioinformatics, vol. 34, no. 15, pp. 2642–2648, 03 2018

    work page 2018

  5. [5]

    Structure-based protein function prediction using graph convolutional networks,

    V . Gligorijevi´c, P. D. Renfrew, T. Kosciolek, J. K. Leman, D. Berenberg, T. Vatanen, C. Chandler, B. C. Taylor, I. M. Fisk, H. Vlamakis, R. J. Xavier, R. Knight, K. Cho, and R. Bonneau, “Structure-based protein function prediction using graph convolutional networks,” Nature Communications, vol. 12, no. 1, p. 3168, May 2021

    work page 2021

  6. [6]

    Comparison of descriptor spaces for chemical compound retrieval and classification,

    N. Wale and G. Karypis, “Comparison of descriptor spaces for chemical compound retrieval and classification,” in Sixth International Conference on Data Mining (ICDM’06) , 2006, pp. 678–689

    work page 2006

  7. [7]

    Moleculenet: A benchmark for molecular machine learning,

    Z. Wu, B. Ramsundar, E. N. Feinberg, J. Gomes, C. Geniesse, A. S. Pappu, K. Leswing, and V . Pande, “Moleculenet: A benchmark for molecular machine learning,” pp. 513–530, 2017

    work page 2017

  8. [8]

    Utilizing graph machine learning within drug discovery and development,

    T. Gaudelet, B. Day, A. R. Jamasb, J. Soman, C. Regep, G. Liu, J. B. R. Hayter, R. Vickers, C. Roberts, J. Tang, D. Roblin, T. L. Blundell, M. M. Bronstein, and J. P. Taylor-King, “Utilizing graph machine learning within drug discovery and development,” Briefings in Bioinformatics , vol. 22, no. 6, 05 2021

    work page 2021

Show all 64 references
  1. [9]

    Could graph neural networks learn better molecular representation for drug discovery? a comparison study of descriptor- based and graph-based models

    D. Jiang, Z. Wu, C. Hsieh, G. Chen, B. Liao, Z. Wang, C. Shen, D. Cao, J. Wu, and T. Hou, “Could graph neural networks learn better molecular representation for drug discovery? a comparison study of descriptor- based and graph-based models.” J Cheminform, vol. 13, no. 1, 02 2021

    2021

  2. [10]

    Tactilegcn: A graph convolutional network for predicting grasp stability with tactile sensors,

    A. Garcia-Garcia, B. S. Zapata-Impata, S. Orts, P. Gil, and J. G. Rodr´ıguez, “Tactilegcn: A graph convolutional network for predicting grasp stability with tactile sensors,” 2019 International Joint Conference on Neural Networks (IJCNN) , pp. 1–8, 2019

    2019

  3. [11]

    Towards federated graph learning for collaborative financial crimes detection,

    T. Suzumura, Y . Zhou, N. Barcardo, G. Ye, K. Houck, R. Kawahara, A. Anwar, L. L. Stavarache, D. Klyashtorny, H. Ludwig, and K. Bhaskaran, “Towards federated graph learning for collaborative financial crimes detection,” CoRR, vol. abs/1909.12946, 2019. [Online]. Available: htt...

    1909 arXiv

  4. [12]

    Fedgnn: Federated graph neural network for privacy-preserving recommendation,

    C. Wu, F. Wu, Y . Cao, Y . Huang, and X. Xie, “Fedgnn: Federated graph neural network for privacy-preserving recommendation,” 2021

    2021

  5. [13]

    Subgraph federated learning with missing neighbor generation,

    K. Zhang, C. Yang, X. Li, L. Sun, and S. M. Yiu, “Subgraph federated learning with missing neighbor generation,” 2021

    2021

  6. [14]

    Federated myopic community detection with one-shot communication,

    C. Ke and J. Honorio, “Federated myopic community detection with one-shot communication,” 2021

    2021

  7. [15]

    Communication-efficient learning of deep networks from decentralized data,

    H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in AISTATS, 2017

    2017

  8. [16]

    Federated graph classification over non-iid graphs,

    H. Xie, J. Ma, L. Xiong, and C. Yang, “Federated graph classification over non-iid graphs,” ArXiv, vol. abs/2106.13423, 2021

    2021 arXiv

  9. [17]

    Facing small and biased data dilemma in drug discovery with enhanced federated learning approaches,

    Z. Xiong, Z. Cheng, X. Lin, C. Xu, X. Liu, D. Wang, X. Luo, Y . Zhang, H. Jiang, N. Qiao, and M. Zheng, “Facing small and biased data dilemma in drug discovery with enhanced federated learning approaches,” Science China. Life sciences , vol. 65, no. 3, p. 529–539, 2022

    2022

  10. [18]

    Nf-gnn: network flow graph neural networks for malware detection and classification,

    J. Busch, A. Kocheturov, V . Tresp, and T. Seidl, “Nf-gnn: network flow graph neural networks for malware detection and classification,” in 33rd International Conference on Scientific and Statistical Database Management, 2021, pp. 121–132

    2021

  11. [19]

    Infograph: Unsupervised and semi-supervised graph-level representation learning via mutual information maximization,

    F.-Y . Sun, J. Hoffman, V . Verma, and J. Tang, “Infograph: Unsupervised and semi-supervised graph-level representation learning via mutual information maximization,” in International Conference on Learning Representations, 2020

    2020

  12. [20]

    Density functional theory,

    R. G. Parr, “Density functional theory,” Annual Review of Physical Chemistry, vol. 34, no. 1, pp. 631–656, 1983

    1983

  13. [21]

    Semi-supervised classification with graph convolutional networks,

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

    2017

  14. [22]

    How powerful are graph neural networks?

    K. Xu, W. Hu, J. Leskovec, and S. Jegelka, “How powerful are graph neural networks?” in International Conference on Learning Representations, 2019. [Online]. Available: https://openreview.net/ forum?id=ryGs6iA5Km

    2019

  15. [23]

    An end- to-end deep learning architecture for graph classification,

    M. Zhang, Z. Cui, M. Neumann, and Y . Chen, “An end- to-end deep learning architecture for graph classification,” ser. AAAI’18/IAAI’18/EAAI’18, 2018

    2018

  16. [24]

    Hierarchical graph representation learning with differentiable pooling,

    R. Ying, J. You, C. Morris, X. Ren, W. L. Hamilton, and J. Leskovec, “Hierarchical graph representation learning with differentiable pooling,” in Proceedings of the 32nd International Conference on Neural Informa- tion Processing Systems , ser. NIPS’18. Curran Associates Inc.,...

    2018

  17. [25]

    graph2vec: Learning distributed representations of graphs,

    A. Narayanan, M. Chandramohan, R. Venkatesan, L. Chen, Y . Liu, and S. Jaiswal, “graph2vec: Learning distributed representations of graphs,” arXiv preprint arXiv:1707.05005 , 2017

    2017 arXiv

  18. [26]

    Strategies for pre-training graph neural networks,

    W. Hu, B. Liu, J. Gomes, M. Zitnik, P. Liang, V . Pande, and J. Leskovec, “Strategies for pre-training graph neural networks,” arXiv preprint arXiv:1905.12265, 2019

    1905 arXiv

  19. [27]

    Biological network comparison using graphlet degree distri- bution,

    N. Pr ˇzulj, “Biological network comparison using graphlet degree distri- bution,” Bioinformatics, vol. 23, no. 2, pp. e177–e183, 2007

    2007

  20. [28]

    Graph invariant kernels,

    F. Orsini, P. Frasconi, and L. De Raedt, “Graph invariant kernels,” in Twenty-Fourth International Joint Conference on Artificial Intelligence , 2015

    2015

  21. [29]

    Contrastive multi-view representa- tion learning on graphs,

    K. Hassani and A. H. Khasahmadi, “Contrastive multi-view representa- tion learning on graphs,” in Proceedings of International Conference on Machine Learning, 2020, pp. 3451–3461

    2020

  22. [30]

    A simple framework for contrastive learning of visual representations,

    T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, “A simple framework for contrastive learning of visual representations,” in International conference on machine learning . PMLR, 2020, pp. 1597–1607

    2020

  23. [31]

    Learning word embeddings efficiently with noise-contrastive estimation,

    A. Mnih and K. Kavukcuoglu, “Learning word embeddings efficiently with noise-contrastive estimation,” Advances in neural information pro- cessing systems, vol. 26, 2013

    2013

  24. [32]

    Model-contrastive federated learning,

    Q. Li, B. He, and D. Song, “Model-contrastive federated learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021

    2021

  25. [33]

    Fedbe: Making bayesian model ensemble applicable to federated learning,

    H.-Y . Chen and W.-L. Chao, “Fedbe: Making bayesian model ensemble applicable to federated learning,” arXiv preprint arXiv:2009.01974 , 2020

    2009 arXiv

  26. [34]

    Federated unsupervised representation learning,

    F. Zhang, K. Kuang, Z. You, T. Shen, J. Xiao, Y . Zhang, C. Wu, Y . Zhuang, and X. Li, “Federated unsupervised representation learning,” arXiv preprint arXiv:2010.08982 , 2020

    2010 arXiv

  27. [35]

    Ensemble distillation for robust model fusion in federated learning,

    T. Lin, L. Kong, S. U. Stich, and M. Jaggi, “Ensemble distillation for robust model fusion in federated learning,” in Advances in Neural Information Processing Systems, H. Larochelle, M. Ranzato, R. Hadsell, M. Balcan, and H. Lin, Eds., vol. 33. Curran Associates, Inc., 2020, ...

    2020

  28. [36]

    Comparing kullback- leibler divergence and mean squared error loss in knowledge distilla- tion,

    T. Kim, J. Oh, N. Kim, S. Cho, and S.-Y . Yun, “Comparing kullback- leibler divergence and mean squared error loss in knowledge distilla- tion,” in Proceedings of the Twenty-Eighth International Joint Confer- ence on Artificial Intelligence, IJCAI-21 . International Joint Conf...

  29. [37]

    Fast subtree kernels on graphs,

    N. Shervashidze and K. Borgwardt, “Fast subtree kernels on graphs,” in Advances in Neural Information Processing Systems , Y . Bengio, D. Schuurmans, J. Lafferty, C. Williams, and A. Culotta, Eds., vol. 22. Curran Associates, Inc., 2009

    2009

  30. [38]

    Graph kernels,

    S. Vishwanathan, N. N. Schraudolph, R. Kondor, and K. M. Borgwardt, “Graph kernels,” Journal of Machine Learning Research , vol. 11, no. 40, pp. 1201–1242, 2010. [Online]. Available: http: //jmlr.org/papers/v11/vishwanathan10a.html

    2010

  31. [39]

    Deep graph kernels,

    P. Yanardag and S. Vishwanathan, “Deep graph kernels,” in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , ser. KDD ’15. New York, NY , USA: Association for Computing Machinery, 2015, p. 1365–1374

    2015

  32. [40]

    Node, motif and subgraph: Leveraging network functional blocks through structural convolution,

    C. Yang, M. Liu, V . W. Zheng, and J. Han, “Node, motif and subgraph: Leveraging network functional blocks through structural convolution,” in 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) , 2018, pp. 47–52

    2018

  33. [41]

    Learning deep representations by mutual information estimation and maximization,

    D. Hjelm, A. Fedorov, S. Lavoie-Marchildon, K. Grewal, P. Bachman, A. Trischler, and Y . Bengio, “Learning deep representations by mutual information estimation and maximization,” in ICLR 2019, 2019

    2019

  34. [42]

    Contrastive multiview coding,

    Y . Tian, D. Krishnan, and P. Isola, “Contrastive multiview coding,” in Computer Vision – ECCV 2020 , 2020

    2020

  35. [43]

    Improved deep metric learning with multi-class n-pair loss objective,

    K. Sohn, “Improved deep metric learning with multi-class n-pair loss objective,” in Advances in Neural Information Processing Systems, 2016

    2016

  36. [44]

    Unsupervised feature learning via non-parametric instance discrimination,

    Z. Wu, Y . Xiong, S. Yu, and D. Lin, “Unsupervised feature learning via non-parametric instance discrimination,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , 2018

    2018

  37. [45]

    Gcc: Graph contrastive coding for graph neural network pre- training,

    J. Qiu, Q. Chen, Y . Dong, J. Zhang, H. Yang, M. Ding, K. Wang, and J. Tang, “Gcc: Graph contrastive coding for graph neural network pre- training,” ser. KDD ’20, 2020, p. 1150–1160

    2020

  38. [46]

    Collaborative unsupervised visual representation learning from decentralized data,

    W. Zhuang, X. Gan, Y . Wen, S. Zhang, and S. Yi, “Collaborative unsupervised visual representation learning from decentralized data,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 4912–4921

    2021

  39. [47]

    Fedcv: a federated learning framework for diverse computer vision tasks,

    C. He, A. D. Shah, Z. Tang, D. F. N. Sivashunmugam, K. Bhogaraju, M. Shimpi, L. Shen, X. Chu, M. Soltanolkotabi, and S. Avestimehr, “Fedcv: a federated learning framework for diverse computer vision tasks,” arXiv preprint arXiv:2111.11066 , 2021

    2021 arXiv

  40. [48]

    Spreadgnn: Serverless multi-task federated learning for graph neural networks,

    C. He, E. Ceyani, K. Balasubramanian, M. Annavaram, and S. Aves- timehr, “Spreadgnn: Serverless multi-task federated learning for graph neural networks,” 2021

    2021

  41. [49]

    Born again neural networks,

    T. Furlanello, Z. Lipton, M. Tschannen, L. Itti, and A. Anandkumar, “Born again neural networks,” in International Conference on Machine Learning. PMLR, 2018, pp. 1607–1616

    2018

  42. [50]

    Understanding and improving knowledge distillation,

    J. Tang, R. Shivanna, Z. Zhao, D. Lin, A. Singh, E. H. Chi, and S. Jain, “Understanding and improving knowledge distillation,” arXiv preprint arXiv:2002.03532, 2020

    2002 arXiv

  43. [51]

    Diffusion improves graph learning,

    J. Klicpera, S. Weißenberger, and S. G ¨unnemann, “Diffusion improves graph learning,” arXiv preprint arXiv:1911.05485 , 2019

    1911 arXiv

  44. [52]

    Representation learning on graphs with jumping knowledge networks,

    K. Xu, C. Li, Y . Tian, T. Sonobe, K.-i. Kawarabayashi, and S. Jegelka, “Representation learning on graphs with jumping knowledge networks,” in International conference on machine learning . PMLR, 2018, pp. 5453–5462

    2018

  45. [53]

    Protein function prediction via graph kernels,

    K. M. Borgwardt, C. S. Ong, S. Sch ¨onauer, S. V . N. Vishwanathan, A. J. Smola, and H.-P. Kriegel, “Protein function prediction via graph kernels,” Bioinformatics, vol. 21, 06 2005

    2005

  46. [54]

    Spline-fitting with a genetic algorithm: a method for developing classification structure- activity relationships,

    J. J. Sutherland, L. A. O’Brien, and D. F. Weaver, “Spline-fitting with a genetic algorithm: a method for developing classification structure- activity relationships,” J. Chem. Inf. Comput. Sci. , vol. 43, no. 6, pp. 1906–1915, Nov. 2003

    1906

  47. [55]

    Comparison of descriptor spaces for chemical compound retrieval and classification,

    N. Wale, I. A. Watson, and G. Karypis, “Comparison of descriptor spaces for chemical compound retrieval and classification,” Knowledge and Information Systems , vol. 14, no. 3, pp. 347–375, Mar 2008

    2008

  48. [56]

    Tudataset: A collection of benchmark datasets for learning with graphs,

    C. Morris, N. M. Kriege, F. Bause, K. Kersting, P. Mutzel, and M. Neu- mann, “Tudataset: A collection of benchmark datasets for learning with graphs,” arXiv preprint arXiv:2007.08663 , 2020

    2007 arXiv

  49. [57]

    Fast graph representation learning with PyTorch Geometric,

    M. Fey and J. E. Lenssen, “Fast graph representation learning with PyTorch Geometric,” in ICLR Workshop on Representation Learning on Graphs and Manifolds , 2019

    2019

  50. [58]

    Decoupled weight decay regularization,

    I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” 2019

    2019

  51. [59]

    Scikit-learn: Machine learning in Python,

    F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V . Dubourg, J. Vander- plas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duch- esnay, “Scikit-learn: Machine learning in Python,” Journal of Machine ...

    2011

  52. [60]

    Federated optimization in heterogeneous networks,

    T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V . Smith, “Federated optimization in heterogeneous networks,” 2020

    2020

  53. [61]

    A metric for distributions with applications to image databases,

    Y . Rubner, C. Tomasi, and L. J. Guibas, “A metric for distributions with applications to image databases,” in Sixth international conference on computer vision (IEEE Cat. No. 98CH36271) . IEEE, 1998, pp. 59–66

    1998

  54. [62]

    Federated visual classification with real-world data distribution,

    T.-M. H. Hsu, H. Qi, and M. Brown, “Federated visual classification with real-world data distribution,” in European Conference on Computer Vision. Springer, 2020, pp. 76–92

    2020

  55. [63]

    A k-means clustering algorithm,

    J. A. Hartigan and M. A. Wong, “A k-means clustering algorithm,” Journal of the Royal Statistical Society. Series C (Applied Statistics) , vol. 28, no. 1, pp. 100–108, 1979

    1979

  56. [64]

    Accelerating t-sne using tree-based algorithms,

    L. Van Der Maaten, “Accelerating t-sne using tree-based algorithms,” J. Mach. Learn. Res. , vol. 15, no. 1, p. 3221–3245, Jan. 2014

    2014

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.