Pith. sign in

REVIEW 3 major objections 6 minor 42 references

Graph Contrastive Learning versus Untrained Baselines: The Role of Dataset Size

T0 review · 3 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read On ogbg-molhiv, GraphCL trails untrained GNN and handcrafted baselines below roughly 4,000 training graphs, then overtakes them by about 2% ROC-AUC before its advantage plateaus.

desk verdict Useful TU benchmark study, but the OGBG-MOLHIV crossover claim is undercut by a training-compute confound that needs fixing before the scaling result can be trusted. read the letter →

arxiv 2509.01541 v1 pith:6Y27XHQJ submitted 2025-09-01 cs.LG cond-mat.soft

Smayan Khanna , Doruk Efe Gökmen , Risi Kondor , Vincenzo Vitelli This is my paper
classification cs.LGcond-mat.soft
keywords graphcontrastivelearninguntrainedbaselinesdatasetsizescalingbehaviorclassificationogbg-molhivsyntheticdatasetsself-supervised
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 asks whether graph contrastive learning (GCL) actually improves over doing nothing: randomly initialized graph encoders, fixed handcrafted graph statistics, or simple neural nets. It argues the answer depends on dataset size. On small TU benchmarks, these untrained or handcrafted baselines often match or beat GCL. On the much larger ogbg-molhiv dataset, GraphCL trails both baselines below roughly 4,000 training graphs, then overtakes them by about 2% ROC-AUC, with the gap plateauing rather than widening as data grows. The paper concludes that dataset size must be a standard axis of GCL evaluation and that small benchmarks are not enough to demonstrate a contrastive advantage.

What carries the argument

The load-bearing setup is the embedding–probe evaluation pipeline: a GNN encoder is pretrained contrastively (or left untrained), frozen, and its graph embeddings are fed to a linear probe. This is paired with systematic subsampling of the training set at fractions from 1% to 100% on ogbg-molhiv, plus a handcrafted-statistics baseline built from node count, mean degree, and a degree histogram. On synthetic data, the paper uses a generator that attaches label-defining motif graphs to random background trees, with a style multiplier controlling task difficulty and dataset size varied independently. Together these components isolate how much of GCL's advantage comes from contrastive pretraining

What would settle it

Retune GraphCL's hyperparameters independently at each subsample fraction of ogbg-molhiv, for example with a per-fraction random search, and compare the crossover against the fixed-hyperparameter curve; a substantially different or absent crossover would show that the dataset-size effect is partly an artifact of tuning. A second check is to repeat the scaling experiment on another large molecular graph dataset with rich node features: if untrained GNNs stay competitive at large scale there, the ~2% advantage is dataset-specific rather than a general property of contrastive learning.

Watch

Extended reading notes

Core claim

The paper's central result is a crossover in relative performance. On ogbg-molhiv, using the official scaffold split, GraphCL pretrained on a training subset performs below both an untrained GINE encoder and the handcrafted-statistics baseline when the training set has up to a few thousand graphs; beyond about 4,000–4,500 graphs GraphCL pulls ahead, but only by roughly 2% ROC-AUC, and this margin stops growing even when training data is increased to the full ~33,000 graphs. On seven TU datasets, the same untrained GNN, a MolFingerprint MLP, and handcrafted statistics rival or beat several established GCL methods, with every GCL method losing on at least one dataset. On synthetic motif-vs-bac

Load-bearing premise

The crossover result rests on the assumption that GraphCL's hyperparameters, chosen once on half the training data, work equally well for every smaller subsample; if those settings are not appropriate at small scale, the point where GCL overtakes the baselines could shift or disappear.

Editorial extensions

If this is right

  • Below a few thousand graphs, GCL gains can be reproduced or beaten by an untrained GNN or handcrafted statistics, so small benchmarks should include such baselines before claiming contrastive success.
  • On ogbg-molhiv, adding data beyond the crossover does not widen GCL's advantage; the gap plateaus at about 2% ROC-AUC, so data scaling alone is not sufficient to justify GCL.
  • The same encoder architecture, evaluated with an embedding probe, makes the untrained GNN a strong random-feature baseline at scale, since its probe can exploit more labels.
  • Dataset size should be reported and varied when comparing GCL methods; crossover points are dataset-dependent but non-trivial relative to standard TU dataset sizes.
  • GCL algorithms need designs that avoid performance plateaus at large scale.

Reading between the lines

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

  • Because GraphCL's hyperparameters were selected on a 50% training fraction and then reused at all fractions, the exact ~4k crossover is likely tunable; per-fraction tuning could shrink GCL's early disadvantage or move the threshold.
  • The crossover location probably depends on how informative node features are, not just dataset size: untrained GNNs are much weaker on synthetic constant-feature graphs than on molecular graphs with rich attributes, so richer features may push crossovers to larger N.
  • A practical evaluation protocol would plot accuracy-versus-data curves against untrained and handcrafted baselines and report the crossover point, rather than single accuracy numbers; the paper's data support this protocol even though the authors only frame it as informative.
  • If log-linear scaling continues on larger datasets, GCL's absolute gain over untrained encoders may remain bounded on tasks where untrained features also improve; methods that break this log scaling would be the ones that matter.
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. The paper evaluates graph contrastive learning (GCL) against untrained and handcrafted baselines across three settings: seven TU benchmark datasets, a subsampled OGBG-MOLHIV scaling experiment, and a synthetic motif dataset with controlled difficulty. On TU datasets, simple baselines (untrained GIN, MolFingerprint, handcrafted degree statistics) are competitive with or superior to InfoGraph, GraphCL, JOAO, and SimGRACE under the standard embedding-SVM protocol. On OGBG-MOLHIV, GraphCL trained with a fixed 100-epoch budget and hyperparameters tuned at 50% data lags below an untrained GINE and handcrafted statistics at small training fractions, crosses over around 4k graphs, and then plateaus at roughly 2% ROC-AUC gain. On synthetic data, both GCL and untrained encoders improve approximately logarithmically with dataset size while handcrafted statistics saturate, with difficulty-dependent gaps. The paper concludes that dataset size is a critical axis for GCL evaluation.

Significance. If correct, the paper's central negative result—small benchmarks and even moderately sized molecular datasets do not reliably demonstrate GCL's advantage—would be a useful corrective to evaluation practice. The manuscript has concrete strengths: the TU comparison is reported in a full table with multiple seeds and ablations; the handcrafted baseline is simple and transparent; the synthetic generator allows controlled variation of difficulty and size; and the authors quote prior numbers where protocols match. The crossover and plateau claims are falsifiable and, if they survive robustness checks, would be an important contribution. However, the OGBG-MOLHIV scaling experiment is built on two protocol choices—fixed epoch count and hyperparameters transferred from the 50% fraction—that both act against GraphCL at small scales. Because these choices are not tested, the main quantitative claim is currently under-supported.

major comments (3)
  1. [Section 3, 'Scaling setup'; App. D] Fixed EPOCHS=100 and BATCH=256 across all subsample fractions confound dataset size with the number of contrastive updates. At 1% (~411 graphs) there are ~2 batches per epoch, so ~200 total updates; at 100% (~41k graphs) there are ~16,100 updates, while the untrained baseline receives zero updates. The observed small-scale GCL deficit and the crossover near 4k graphs may therefore reflect underfitting from an extremely small optimization budget rather than a property of dataset size itself. This confound is not acknowledged. Please add a control that trains GraphCL for a fixed number of total updates across fractions (e.g., by scaling epochs inversely with fraction), and ideally report results for two or three training budgets, to show whether the crossover persists.
  2. [Section 3, App. D, Table 4] Hyperparameters (augmentation strategy, hidden dimension, learning rate) were selected by grid search on a 50% subset and then applied unchanged to all fractions from 1% to 100%. Optimal hyperparameters in the low-data regime are likely different (e.g., different capacity, regularization, or augmentation drop rates), so this transfer assumption can bias small fractions against GraphCL. The crossover location is directly sensitive to this choice. Please run per-fraction tuning (even a coarse version) or at least a sensitivity ablation at the 1%, 5%, and 20% fractions around the Table 4 settings, and report whether the crossover shifts.
  3. [Section 4, Fig. 3; Conclusion] The claim that 'GCL accuracy approximately scales with the logarithm of the number of graphs' is asserted from visual inspection of four points per condition. No slopes, standard errors, or model-comparison statistics are reported. If this scaling statement is a main conclusion, it needs a quantitative fit (e.g., accuracy vs. log N with confidence bands) or should be explicitly softened to 'roughly linear over the tested range.' As written, the abstract's logarithmic-scaling claim is stronger than the evidence presented.
minor comments (6)
  1. [Throughout] OGBG-MOLHIV is capitalized inconsistently; the abstract uses 'ogbg-molhiv'. Please standardize.
  2. [Section 3, Scaling setup] Typo: 'samef-subset' should be 'same f-subset'.
  3. [App. C.1] For un-attributed TU datasets (COLLAB, IMDB-BINARY, REDDIT-BINARY), please specify how node features are constructed for the GIN encoder and MolFingerprint (e.g., degree features or constant vectors). This affects reproducibility of the untrained baselines.
  4. [Table 2] The abbreviation RDT-B in the table header is undefined; spell out REDDIT-BINARY or define it in the caption.
  5. [App. D, Table 4] The text says hidden dimension is searched over {32, 64}; Table 4 lists HIDDEN=32 and PROJ=64. Clarify whether the projection dimension was also searched and whether Table 4 reports the final choice for both.
  6. [Figure 2] The right-panel labels 'Δ vs baseline' and 'Δ vs handcrafted' are clear, but the x-axis tick values (300, 1000, 5000, 30000) omit the 1% and 2% points; consider marking all sampled fractions for transparency.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: empirical benchmark against external datasets and prior baselines; central claims are observations, not reductions to fitted inputs.

full rationale

The paper's derivation chain is an empirical comparison: GCL methods and baselines are evaluated on external TU datasets, OGBG-MOLHIV, and a synthetic generator from prior work. No parameter is fit to the target result: the handcrafted statistics baseline is a fixed descriptor (node count, mean degree, degree histogram) with an SVM probe; the untrained GNN is a frozen random GIN/GINE; the crossover is read off measured ROC-AUC curves, not derived from an equation involving the fitted values. Hyperparameters for GraphCL are tuned on a 50% OGBG subset and transferred to all fractions; this is a methodological assumption about transferability, and the fixed epoch count confounds dataset size with number of optimizer updates, but these are experimental-validity concerns, not circular reductions. There are no self-citations by the authors, and no invoked uniqueness theorem or ansatz that smuggles the conclusion. The conclusion explicitly flags open questions ('why crossovers occur, when gains saturate'), so the paper does not present the crossover as a forced mathematical consequence. Score 0.

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

The central claims rest on standard evaluation-protocol assumptions inherited from the GCL literature, author-designed baselines, and one untested transfer assumption: that hyperparameters tuned at the 50% fraction of OGBG-MOLHIV are appropriate for all subsample sizes. No new physical or mathematical entities are introduced.

free parameters (4)
  • GraphCL hyperparameters on OGBG-MOLHIV = EPOCHS=100, BATCH=256, LR=1e-3, WD=1e-5, TEMP=0.2, PROJ=64, HIDDEN=32, edge-drop p=0.10
    Tuned via grid search on the 50% training fraction only; the same configuration is used for all subsample fractions. The crossover location is sensitive to this choice.
  • GraphCL hyperparameters on synthetic datasets = EPOCHS=50, BATCH=256, HIDDEN=32, LAYERS=3, LR=1e-3, PROJ=32, ND+ED p=0.15
    Fixed settings applied across all dataset sizes and difficulties; no per-size tuning, which may affect the scaling curves.
  • InfoGraph projection dimension on synthetic datasets = PROJ=2
    Fixed setting; unusually small projection, not tuned across sizes.
  • Handcrafted statistics descriptor = node count + mean degree + 5-bin degree histogram, SVM C selected by cross-validation
    Author-designed feature set; ablations in Table 3 show the degree histogram dominates, but the descriptor itself is a design choice that influences the baseline's strength.
assumptions (5)
  • domain assumption The embedding-probe protocol (frozen encoder + linear classifier) measures representation quality for GCL.
    Adopted from prior work [24,27]; all comparisons use this protocol, so conclusions are conditional on its validity.
  • domain assumption A randomly initialized, frozen GNN is a meaningful untrained baseline.
    Used throughout; the paper does not test sensitivity to initialization distribution or seed count beyond 5 seeds.
  • domain assumption OGB's official scaffold split remains valid after subsampling the training set.
    Sec. 3; subsampling may alter label balance and difficulty per fraction, which is not analyzed.
  • domain assumption Trivedi et al.'s synthetic generator's style multiplier S controls task difficulty.
    Sec. 4; the paper uses S in {2,4,6} as easy/medium/hard without independently re-validating difficulty.
  • ad hoc to paper Hyperparameters tuned on the 50% fraction of OGBG-MOLHIV transfer to all other fractions.
    App D; this is load-bearing for the crossover claim, is untested, and is not acknowledged as an assumption.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph Contrastive Learning versus Untrained Baselines: The Role of Dataset Size." pith.science (2026). https://pith.science/paper/6Y27XHQJ

@misc{pith2026250901541,
  author       = {Pith},
  title        = {Pith review of: Graph Contrastive Learning versus Untrained Baselines: The Role of Dataset Size},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6Y27XHQJ}},
  note         = {Machine review of arXiv:2509.01541}
}
read the original abstract

Graph Contrastive Learning (GCL) has emerged as a leading paradigm for self-supervised learning on graphs, with strong performance reported on standardized datasets and growing applications ranging from genomics to drug discovery. We ask a basic question: does GCL actually outperform untrained baselines? We find that GCL's advantage depends strongly on dataset size and task difficulty. On standard datasets, untrained Graph Neural Networks (GNNs), simple multilayer perceptrons, and even handcrafted statistics can rival or exceed GCL. On the large molecular dataset ogbg-molhiv, we observe a crossover: GCL lags at small scales but pulls ahead beyond a few thousand graphs, though this gain eventually plateaus. On synthetic datasets, GCL accuracy approximately scales with the logarithm of the number of graphs and its performance gap (compared with untrained GNNs) varies with respect to task complexity. Moving forward, it is crucial to identify the role of dataset size in benchmarks and applications, as well as to design GCL algorithms that avoid performance plateaus.

Figures

Figures reproduced from arXiv: 2509.01541 by the authors.

Figure 1
Figure 1. Graph classification accuracy difference (%) between GCL and baselines. Rows show representative GCL methods, columns the TU datasets. Left: GCL vs the best-performing simple baseline for that dataset. Right: GCL vs. the handcrafted statistics baseline. Negative values (blue) indicate that the baseline outperforms the GCL method, while positive values (red) favor GCL. protocol [24], which evaluates representation qu… view at source ↗
Figure 2
Figure 2. Scaling on OGBG-MOLHIV (Log Scale). Top: GraphCL vs. an untrained GINE. Bottom: GraphCL vs. the handcrafted statistics baseline. In each row, the left panel shows test ROC-AUC vs. number of training graphs with shaded ±1σ over seeds; the right panel shows ∆ (GraphCL−baseline, in percentage points) with the vertical dotted line marking the approximate crossover where ∆ = 0. Positive values favor GraphCL. f-subset for… view at source ↗
Figure 3
Figure 3. Scaling on Synthetic Datasets. Top: GraphCL. Test accuracy vs. total number of graphs (log scale) for S ∈ {2, 4, 6}; bands show ±1σ over seeds; handcrafted baseline shown as dotted line. Bottom: ∆ heatmaps for GraphCL—Untrained and GraphCL—Handcrafted Statistics baseline across style multiplier S and dataset size (shared color scale; positive values favor GraphCL). A notable pattern is that both GCL and the untraine… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Scaling on Synthetic Datasets. Top: InfoGraph. Test accuracy vs. total number of graphs (log scale) for S ∈ {2, 4, 6}; bands show ±1σ over seeds; handcrafted baseline shown as dotted line. Bottom: ∆ heatmaps for InfoGraph — Untrained and InfoGraph — Handcrafted across …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

42 extracted references · 30 canonical work pages

  1. [1]

    Towards graph contrastive learning: A survey and beyond, 2024

    Wei Ju, Yifan Wang, Yifang Qin, Zhengyang Mao, Zhiping Xiao, Junyu Luo, Junwei Yang, Yiyang Gu, Dongjie Wang, Qingqing Long, Siyu Yi, Xiao Luo, and Ming Zhang. Towards graph contrastive learning: A survey and beyond, 2024. 1, 8

    work page 2024

  2. [2]

    Prediction of multi-relational drug–gene interaction via dynamic hypergraph contrastive learning

    Wen Tao, Yuansheng Liu, Xuan Lin, Bosheng Song, and Xiangxiang Zeng. Prediction of multi-relational drug–gene interaction via dynamic hypergraph contrastive learning. Briefings in Bioinformatics, 24(6):bbad371, 10 2023. 1

    work page 2023

  3. [3]

    Predicting drug–target binding affinity with cross-scale graph contrastive learning

    Jingru Wang, Yihang Xiao, Xuequn Shang, and Jiajie Peng. Predicting drug–target binding affinity with cross-scale graph contrastive learning. Briefings in Bioinformatics, 25(1):bbad516, 01 2024. 1

    work page 2024

  4. [4]

    Deep single-cell rna-seq data clustering with graph prototypical contrastive learning

    Junseok Lee, Sungwon Kim, Dongmin Hyun, Namkyeong Lee, Yejin Kim, and Chanyoung Park. Deep single-cell rna-seq data clustering with graph prototypical contrastive learning. Bioinformatics, 39(6):btad342, 05 2023. 1

    work page 2023

  5. [5]

    scgcl: an imputation method for scrna-seq data based on graph contrastive learning

    Zehao Xiong, Jiawei Luo, Wanwan Shi, Ying Liu, Zhongyuan Xu, and Bo Wang. scgcl: an imputation method for scrna-seq data based on graph contrastive learning. Bioinformatics, 39 (3):btad098, 02 2023. 1

    work page 2023

  6. [6]

    Towards robust false information detection on social networks with contrastive learning

    Guanghui Ma, Chunming Hu, Ling Ge, Junfan Chen, Hong Zhang, and Richong Zhang. Towards robust false information detection on social networks with contrastive learning. In Proceedings of the 31st ACM International Conference on Information & Knowledge Management, CIKM ’22, page 1441–1450, New York, NY , USA, 2022. Association for Computing Machinery. ISBN ...

    arXiv 2022

  7. [7]

    Rumor detection on social media with graph adversarial contrastive learning

    Tiening Sun, Zhong Qian, Sujun Dong, Peifeng Li, and Qiaoming Zhu. Rumor detection on social media with graph adversarial contrastive learning. In Proceedings of the ACM Web Conference 2022, WWW ’22, page 2789–2797, New York, NY , USA, 2022. Association for Computing Machinery. ISBN 9781450390965. doi: 10.1145/3485447.3511999. URL https://doi.org/10.1145/...

    arXiv 2022

  8. [8]

    A fair comparison of graph neural networks for graph classification

    Federico Errica, Marco Podda, Davide Bacciu, and Alessio Micheli. A fair comparison of graph neural networks for graph classification. In International Conference on Learning Representations, 2020. 1, 2, 8, 9

    work page 2020

Show all 42 references
  1. [9]

    Kriege, Franka Bause, Kristian Kersting, Petra Mutzel, and Marion Neumann

    Christopher Morris, Nils M. Kriege, Franka Bause, Kristian Kersting, Petra Mutzel, and Marion Neumann. Tudataset: A collection of benchmark datasets for learning with graphs. In ICML 2020 Workshop on Graph Representation Learning and Beyond (GRL+ 2020) , 2020. URL www.graphlea...

    2020

  2. [10]

    Yuzhou Chen, Jose Frias, and Yulia R. Gel. Topogcl: Topological graph contrastive learning. Proceedings of the AAAI Conference on Artificial Intelligence , 38(10):11453–11461, Mar

  3. [11]

    Graph contrastive learning with reinforcement augmentation

    Ziyang Liu, Chaokun Wang, and Cheng Wu. Graph contrastive learning with reinforcement augmentation. In Kate Larson, editor, Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, IJCAI-24, pages 2225–2233. International Joint Confer- ences o...

    2024 doi

  4. [12]

    Multi-scale subgraph con- trastive learning

    Yanbei Liu, Yu Zhao, Xiao Wang, Lei Geng, and Zhitao Xiao. Multi-scale subgraph con- trastive learning. In Edith Elkind, editor, Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, IJCAI-23, pages 2215–2223. International Joint Confer- e...

    2023 doi

  5. [13]

    Graph contrastive learning with cohesive subgraph awareness

    Yucheng Wu, Leye Wang, Xiao Han, and Han-Jia Ye. Graph contrastive learning with cohesive subgraph awareness. In Proceedings of the ACM Web Conference 2024 , WWW ’24, page 629–640, New York, NY , USA, 2024. Association for Computing Machinery. ISBN 9798400701719. doi: 10.1145/...

    2024

  6. [14]

    Khan-gcl: Kolmogorov-arnold network based graph contrastive learning with hard negatives, 2025

    Zihu Wang, Boxun Xu, Hejia Geng, and Peng Li. Khan-gcl: Kolmogorov-arnold network based graph contrastive learning with hard negatives, 2025. URL https://arxiv.org/abs/2505. 15103. 1, 9

    2025

  7. [15]

    Open graph benchmark: Datasets for machine learning on graphs

    Weihua Hu, Matthias Fey, Marinka Zitnik, Yuxiao Dong, Hongyu Ren, Bowen Liu, Michele Catasta, and Jure Leskovec. Open graph benchmark: Datasets for machine learning on graphs. In Advances in Neural Information Processing Systems, 2020. 1, 2, 10

    2020

  8. [16]

    Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S

    Zhenqin Wu, Bharath Ramsundar, Evan N. Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S. Pappu, Karl Leswing, and Vijay Pande. Moleculenet: a benchmark for molecular machine learning. Chem. Sci., 9:513–530, 2018. doi: 10.1039/C7SC02664A. 1, 2, 10

    2018 doi

  9. [17]

    Thiagara- jan

    Puja Trivedi, Ekdeep Singh Lubana, Mark Heimann, Danai Koutra, and Jayaraman J. Thiagara- jan. Analyzing data-centric properties for graph contrastive learning. In Advances in Neural Information Processing Systems, 2022. 1, 3, 8, 11

    2022

  10. [18]

    Aug- mentations in graph contrastive learning: Current methodological flaws & towards better practices

    Puja Trivedi, Ekdeep Singh Lubana, Yujun Yan, Yaoqing Yang, and Danai Koutra. Aug- mentations in graph contrastive learning: Current methodological flaws & towards better practices. In Proceedings of the ACM Web Conference 2022, WWW ’22, page 1538–1549, New York, NY , USA, 202...

    2022

  11. [19]

    A simple yet effective baseline for non-attributed graph classification,

    Chen Cai and Yusu Wang. A simple yet effective baseline for non-attributed graph classification,

  12. [20]

    Benchmarking and analyzing unsupervised network represen- tation learning and the illusion of progress

    Saket Gurukar, Priyesh Vijayan, srinivasan parthasarathy, Balaraman Ravindran, Aakash Srini- vasan, Goonmeet Bajaj, Chen Cai, Moniba Keymanesh, Saravana Kumar, Pranav Maneriker, Anasua Mitra, and Vedang Patel. Benchmarking and analyzing unsupervised network represen- tation le...

    2022

  13. [21]

    Graph contrastive learning with augmentations

    Yuning You, Tianlong Chen, Yongduo Sui, Ting Chen, Zhangyang Wang, and Yang Shen. Graph contrastive learning with augmentations. In Advances in Neural Information Processing Systems, 2020. 1, 2, 8

    2020

  14. [22]

    Graph contrastive learning automated

    Yuning You, Tianlong Chen, Yang Shen, and Zhangyang Wang. Graph contrastive learning automated. In Marina Meila and Tong Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, pages 12121–12...

    2021

  15. [23]

    Jun Xia, Lirong Wu, Jintao Chen, Bozhen Hu, and Stan Z. Li. Simgrace: A simple framework for graph contrastive learning without data augmentation. In Proceedings of the ACM Web Conference 2022, WWW ’22, page 1070–1079, New York, NY , USA, 2022. Association for Computing Machin...

    2022

  16. [24]

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

    Fan-Yun Sun, Jordan Hoffman, Vikas Verma, and Jian Tang. Infograph: Unsupervised and semi-supervised graph-level representation learning via mutual information maximization. In International Conference on Learning Representations, 2020. 1, 2, 8, 9

    2020

  17. [25]

    How powerful are graph neural networks? In International Conference on Learning Representations, 2019

    Keyulu Xu, Weihua Hu, Jure Leskovec, and Stefanie Jegelka. How powerful are graph neural networks? In International Conference on Learning Representations, 2019. 2

    2019

  18. [26]

    Rethinking the effective- ness of graph classification datasets in benchmarks for assessing gnns

    Zhengdao Li, Yong Cao, Kefan Shuai, Yiming Miao, and Kai Hwang. Rethinking the effective- ness of graph classification datasets in benchmarks for assessing gnns. In Kate Larson, editor, Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, ...

  19. [27]

    An empirical study of graph contrastive learning

    Yanqiao Zhu, Yichen Xu, Qiang Liu, and Shu Wu. An empirical study of graph contrastive learning. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021. 2, 8, 10

    2021

  20. [28]

    Strategies for pre-training graph neural networks

    Weihua Hu*, Bowen Liu*, Joseph Gomes, Marinka Zitnik, Percy Liang, Vijay Pande, and Jure Leskovec. Strategies for pre-training graph neural networks. In International Conference on Learning Representations, 2020. URL https://openreview.net/forum?id=HJlWWJSFDH. 2 6 Graph Contra...

    2020

  21. [29]

    URL https://doi.org/10.24963/ijcai.2024/237

    doi: 10.24963/ijcai.2024/237. URL https://doi.org/10.24963/ijcai.2024/237. Main Track. 2, 4, 8

    2024 doi

  22. [30]

    Matthias Fey and Jan E. Lenssen. Fast graph representation learning with PyTorch Geometric. In ICLR Workshop on Representation Learning on Graphs and Manifolds, 2019. 8

    2019

  23. [31]

    Lenssen, and Jure Leskovec

    Matthias Fey, Jinu Sunil, Akihiro Nitta, Rishi Puri, Manan Shah, Blaž Stojanovi ´c, Ramona Bendias, Barghi Alexandria, Vid Kocijan, Zecheng Zhang, Xinwei He, Jan E. Lenssen, and Jure Leskovec. PyG 2.0: Scalable learning on real world graphs. In Temporal Graph Learning Workshop...

    2025

  24. [32]

    Architecture matters: Uncovering implicit mechanisms in graph contrastive learning

    Xiaojun Guo, Yifei Wang, Zeming Wei, and Yisen Wang. Architecture matters: Uncovering implicit mechanisms in graph contrastive learning. In Advances in Neural Information Pro- cessing Systems, 2023. URL https://proceedings.neurips.cc/paper_files/paper/ 2023/file/5acf5a0ee5c17d...

    2023

  25. [33]

    Borgwardt

    Nino Shervashidze, Pascal Schweitzer, Erik Jan van Leeuwen, Kurt Mehlhorn, and Karsten M. Borgwardt. Weisfeiler-lehman graph kernels. Journal of Machine Learning Research, 12(77): 2539–2561, 2011. URL http://jmlr.org/papers/v12/shervashidze11a.html. 9

    2011

  26. [34]

    Vishwanathan

    Pinar Yanardag and S.V .N. Vishwanathan. Deep graph kernels. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , KDD ’15, page 1365–1374, New York, NY , USA, 2015. Association for Computing Machinery. ISBN 9781450336642. doi:...

    2015

  27. [35]

    Representation learning with contrastive predictive coding, 2019

    Aaron van den Oord, Yazhe Li, and Oriol Vinyals. Representation learning with contrastive predictive coding, 2019. URL https://arxiv.org/abs/1807.03748. 8, 10

    2019 arXiv

  28. [36]

    node2vec: Scalable feature learning for networks

    Aditya Grover and Jure Leskovec. node2vec: Scalable feature learning for networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , KDD ’16, page 855–864, New York, NY , USA, 2016. Association for Computing Machinery. ISBN...

    2016

  29. [37]

    graph2vec: Learning distributed representations of graphs, 2017

    Annamalai Narayanan, Mahinthan Chandramohan, Rajasekar Venkatesan, Lihui Chen, Yang Liu, and Shantanu Jaiswal. graph2vec: Learning distributed representations of graphs, 2017. URL https://arxiv.org/abs/1707.05005. 9

    2017 arXiv

  30. [38]

    Aditya Prakash

    Bijaya Adhikari, Yao Zhang, Naren Ramakrishnan, and B. Aditya Prakash. Sub2vec: Feature learning for subgraphs. In Advances in Knowledge Discovery and Data Mining: 22nd Pacific- Asia Conference, PAKDD 2018, Melbourne, VIC, Australia, June 3-6, 2018, Proceedings, Part II, page ...

    2018 doi

  31. [41]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In ICLR,

  32. [2015]

    GCL learns node- or graph-level representations by contrasting positive and negative views generated from the data itself [1]

    10 7 Graph Contrastive Learning vs Untrained Baselines: The Role of Dataset Size A Expanded Related Work Graph contrastive learning. GCL learns node- or graph-level representations by contrasting positive and negative views generated from the data itself [1]. We focus on metho...

  33. [2022]

    URL https://arxiv.org/abs/1811.03508. 1

    arXiv

  34. [2024]

    URL https://ojs.aaai.org/index.php/AAAI/ article/view/29026

    doi: 10.1609/aaai.v38i10.29026. URL https://ojs.aaai.org/index.php/AAAI/ article/view/29026. 1, 9

    doi

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.