Pith. sign in

REVIEW 5 major objections 5 minor 116 references

Towards Automated Self-Supervised Learning for Truly Unsupervised Graph Anomaly Detection

T0 review · 5 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read AutoGAD claims that a label-free score based on the Contrast Score Margin can select SSL hyperparameters for graph anomaly detection at nearly the performance of label-guided tuning.

desk verdict A useful empirical study of label leakage in SSL-based graph anomaly detection, with a pragmatic internal evaluation strategy that mostly beats random hyperparameter choice—but the theory does not cover the actual pseudo-label procedure. read the letter →

arxiv 2501.14694 v2 pith:7YSDGOPU submitted 2025-01-24 cs.LG cs.AI

Zhong Li , Yuhang Wang , Matthijs van Leeuwen This is my paper · ORCID
classification cs.LGcs.AI
keywords graphanomalydetectionself-supervisedlearninghyperparametertuninglabelleakageinternalevaluationcontrastscoremarginunsupervisedmodelselectionneuralnetworks
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 argues that many recent self-supervised graph anomaly detection (GAD) methods are not truly unsupervised: their hyperparameters are chosen either arbitrarily or by looking at ground-truth labels, and the label-guided choices inflate reported AUC. It proposes AutoGAD, a plug-and-play procedure that selects SSL strategies' combination weights and hyperparameters using the Contrast Score Margin, an internal metric computed only from anomaly scores, with no labels. The central claim is that maximizing this margin is equivalent to simultaneously minimizing false positive and false negative rates, and that grid search over this score picks configurations at or near the best achievable AUC on most of the ten algorithms and ten datasets tested. If correct, practitioners can tune SSL-based GAD methods without label leakage, and several published performance numbers need to be re-read as optimistic upper bounds.

What carries the argument

The central object is the modified Contrast Score Margin, $T(f) = \frac{\hat{\mu}_O - \tilde{\mu}_I}{\sqrt{\hat{\delta}^2_O + \tilde{\delta}^2_I}}$, where $\hat{\mu}_O, \hat{\delta}^2_O$ are the mean and variance of the anomaly scores of the $k$ predicted anomalous nodes and $\tilde{\mu}_I, \tilde{\delta}^2_I$ are the mean and variance of the remaining $n-k$ nodes. The statistic measures how far the top-$k$ pseudo-anomalies stand out from the rest, normalized by score spread, and runs in linear time on the score vector. Grid search over a discretized hyperparameter space maximizes $T$ to choose the configuration. The theoretical support uses Cantelli's inequality to bound the probability that a true anomaly falls below its group mean minus a multiple of its standard deviation, and the probability that a normal node rises above its group mean plus a multiple; making the gap between the two means large relative to the standard deviations is therefore the same as shrinking both false-negative and false-positive bounds. The theorem is stated for true anomaly sets, whereas in practice the sets are replaced by pseudo-labels derived from top-$k$ scores.

What would settle it

Take a dataset whose true anomalies are known, hide the labels, and compute the Spearman correlation between the modified Contrast Score Margin in Eq. 3 and true AUC over all grid configurations while varying $k$ from 0.25x to 4x the true anomaly count; if the correlation is near zero or negative for a reasonable value of $k$, the claim that maximizing $T$ approximates maximizing AUC fails on that setting.

Watch

Extended reading notes

Core claim

For a fixed SSL-based graph anomaly detection algorithm, the modified Contrast Score Margin $T(f) = \frac{\hat{\mu}_O - \tilde{\mu}_I}{\sqrt{\hat{\delta}^2_O + \tilde{\delta}^2_I}}$ ranks hyperparameter configurations almost as well as true AUC when the top-$k$ pseudo-anomalies are the $k$ nodes with the highest anomaly scores and the remaining $n-k$ nodes form the pseudo-normal group. The paper claims in Theorem 1, via Cantelli's inequality, that maximizing this margin simultaneously minimizes false positives and false negatives, and it shows empirically that selecting the configuration with the largest $T$ yields AUC close to the maximum over a large grid for most algorithm-dataset pairs. A correctness boundary stated by the paper is that this score should select among hyperparameter settings of the same algorithm with comparable loss scales, not among heterogeneous detectors; the appendix shows it fails at the latter.

Load-bearing premise

The paper assumes the anomaly ratio is roughly known so that $k$, the number of nodes treated as pseudo-anomalies in the score, is near the true number of anomalies; if $k$ is wrong, the top-$k$ set contains normal nodes and the margin being maximized no longer tracks true detection quality.

Editorial extensions

If this is right

  • AutoGAD gives a label-free procedure for choosing augmentation functions, hyperparameters, and combination weights for any SSL-based GAD algorithm that outputs anomaly scores, removing the need for a labeled validation set.
  • Because the selection criterion is computed entirely from scores, the same detector can be adapted per dataset through grid search instead of relying on fixed heuristics that were tuned with labels.
  • Published results of SSL-based GAD baselines that selected hyperparameters using ground-truth labels should be regarded as upper bounds; a truly unsupervised deployment is likely to land lower, sometimes substantially.
  • The Contrast Score Margin is not a universal model selector: the authors show it fails to rank heterogeneous detection algorithms, so it should be used only within one algorithm's hyperparameter space with comparable loss scales.

Reading between the lines

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

  • If the anomaly ratio is not approximately known, the effectiveness of AutoGAD could degrade sharply, because $k$ sets the pseudo-label boundary; a testable extension would feed a label-free estimate of $k$ (e.g., from score gaps or a silhouette-style criterion) into the margin and check whether selection quality is preserved.
  • The same margin-based reasoning might transfer to unsupervised model selection beyond graphs, such as tabular or time-series anomaly detectors, whenever scores from one detector family are comparable; the paper does not claim this extension.
  • The reported sensitivity results imply that even moderate hyperparameter variation can change AUC by 15 to 30 percent for several methods, so an ensemble over several high-margin configurations may be a more robust practical recipe than picking a single best configuration, though the paper only reports single-configuration selection.
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

5 major / 5 minor

Summary. The paper identifies label leakage in SSL-based graph anomaly detection (GAD), where hyperparameters are often tuned using ground-truth labels, and proposes AutoGAD, a label-free internal evaluation strategy based on a modified Contrast Score Margin (CSM). The method selects SSL hyperparameters via grid search using Eq. (3), which measures the margin between the top-k predicted anomalous nodes and all remaining nodes, without using labels. The authors provide a theoretical analysis (Theorem 1) claiming that maximizing CSM is equivalent to minimizing false positive and false negative rates, and report extensive experiments across 10 SSL-based GAD algorithms and 10 datasets showing that CSM-selected hyperparameters beat the median and, in many cells, approach the maximum AUC. The paper also documents that many existing methods report overestimated performance due to label-based tuning.

Significance. The label-leakage documentation is a valuable and timely contribution: the paper gives concrete evidence (Table 1, Appendix B) that SSL-based GAD methods are highly sensitive to SSL hyperparameter choices and that reported performances often come from label-guided tuning. The proposed AutoGAD is a plausible and lightweight approach to label-free hyperparameter selection, and the experimental setup is extensive, including 10 algorithms, 10 datasets, multiple runs, sensitivity analyses, and released code. If the CSM criterion provably tracked AUC across hyperparameter configurations, this would be a practical advance for unsupervised graph anomaly detection. However, the theoretical support does not cover the actual pseudo-label-based metric, and the empirical claim of approximating the maximum AUC is not uniformly supported, so the central claim is not yet established.

major comments (5)
  1. [§5.1.2–§5.1.3] Theorem 1 is proved for the true sets O and I of top-k anomalies and top-k normals, but the method selects hyperparameters with Eq. (3), in which the 'anomaly' set is the top-k pseudo-anomalies obtained from the detector's own scores and the 'normal' set is all remaining n−k objects. The Cantelli bounds in the theorem do not apply when O and I are replaced by these pseudo-labels; Section 5.1.3 itself notes that the pseudo-labels of top-k normals may be unreliable. Since the theorem is the stated theoretical justification for Eq. (3), this is a gap between the guarantee and the procedure actually used.
  2. [§5.1.2, proof of Theorem 1] The proof argues that to keep both bounds small one wants µO−µI large and aδO+bδI small, and then asserts that this is 'equivalent' to maximizing T = (µO−µI)/sqrt(δO^2+δI^2). This is not a formal equivalence: a ratio can increase because the denominator shrinks, without the separate bounds improving, and the proof gives no argument that optimizing the single ratio is equivalent to simultaneously controlling the two inequalities. The theorem as stated is therefore not established, even for true O and I.
  3. [Table 5 and §6.5.2, observation 3] The claim that CSM-selected hyperparameters 'approximately achieve the best possible performance' is not uniformly supported by the reported gains over maximal AUC. Several cells have double-digit losses: CoLA on YelpChi −30.2%, ANEMONE on YelpChi −30.3%, SL-GAD on Facebook −39.4%, GAAN on Facebook −53.7%, and CONAD on YelpChi −26.3%; the averages in the rightmost column are −3.8% to −14.1%. The paper should either qualify the claim, analyze the failure cases, or restrict it to cells where the loss is small.
  4. [§6.5.3, Sensitivity to k] The method requires the number of pseudo-anomalies k, and the paper states that it 'operated under the assumption that the anomaly ratio within a dataset is approximately known.' This is external prior information that weakens the 'truly unsupervised' claim, and it is load-bearing because with a wrong k or a poor score ranking the top-k set is polluted and maximizing Eq. (3) can separate an arbitrary high-scoring cluster from the bulk without improving detection. The sensitivity analysis in Figure 3 covers only CiteSeer, and it does not test the realistic case where the supplied k is far from the true ratio for the algorithms and datasets on which Table 5 shows large losses.
  5. [Appendix F and §5.1.3] Because Eq. (3) is defined from the detector's own anomaly scores, maximizing it could reward score-separation artifacts rather than true detection. The paper provides cross-method evidence in Appendix F that CSM does not track AUC across heterogeneous detectors (Pearson correlations 0.070 on Cora and −0.488 on Amazon), but it does not report the analogous within-algorithm correlation between CSM and AUC over the HP grids used in Tables 3–5. This within-method criterion validity is the direct empirical support needed for the HPO claim; without it, the benchmark results in Tables 3–5 remain the only partial evidence.
minor comments (5)
  1. [Table 2 caption] The caption says 'PubWeb' but the dataset is PubMed; also, 'BlogCataLog' is spelled inconsistently throughout the text and tables.
  2. [Table 6] Table 6 attributes CONAD to 'Zhang et al. (2022)', but the CONAD method is cited and described in the text as Xu et al. (2022b); this citation should be corrected.
  3. [§5.1.3, Eq. (3)] Equation (3) drops the 1/k factor inside the square root of Eq. (2) without comment; since k is a free parameter, the revised metric's scale and k-dependence change and this change should be discussed explicitly.
  4. [§6.5.2] The text says '8 out 10 algorithms' where '8 out of 10' is intended; also, the example citing SL-GAD as highly effective is odd because its average gain over max AUC is −7.5%, worse than several other methods.
  5. [Figure 3] The sensitivity analysis for k is shown only on CiteSeer, while the surrounding text and the 'truly unsupervised' framing suggest broader support; the figure caption or text should clarify that Figure 3 alone does not establish stability across datasets.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reduction found: AutoGAD's internal metric is not fitted from labels, and its effectiveness is benchmarked against independent AUC values.

full rationale

The selection criterion T(f) in Eq. 3 is computed from the detector's own anomaly scores and pseudo-labels, but it is not fitted to ground-truth labels, and the reported AUC numbers are evaluated independently with labels reserved for performance assessment. The largest logical gap is that Theorem 1 is proved for the true anomalous set O and the top-k true normal set I of Eq. 2, whereas the deployed criterion Eq. 3 uses the top-k predicted anomalies and the remaining n-k objects; Section 5.1.3 itself notes that these pseudo-labels may be unreliable. This is a missing-support or overclaim issue, not a circular reduction: Eq. 3 is not defined in terms of the target AUC, and the experiments show substantial deviations from the oracle maximum in several cases (e.g., CoLA on YelpChi -30.2% in Table 5), which is incompatible with an equivalence-by-construction claim. Section 6.5.3 explicitly acknowledges that k assumes approximate knowledge of the anomaly ratio, and Appendix F reports that CSM fails to rank heterogeneous detectors; these are stated limitations rather than hidden inputs that force the claimed conclusion. No load-bearing self-citation or imported uniqueness theorem is used; the only self-citations are contextual related work. The central claim therefore has independent empirical content, and no circular step is established.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No new physical or methodological entities are introduced; AutoGAD is a selection procedure over existing detectors. The main ledger entries are the user-specified k and the pseudo-label fidelity assumption, both of which are load-bearing for the central claim.

free parameters (1)
  • k (number of pseudo-anomalies in CSM) = set to true anomaly count per dataset in experiments (e.g., CiteSeer true ratio 4.5%); user must know anomaly ratio
    T(f) in Eq. 3 is defined on the top-k anomaly scores; the paper assumes this ratio is approximately known (Section 6.5.3) and shows sensitivity when k deviates. This is a hand-specified input, not learned from data.
assumptions (4)
  • standard math Cantelli's inequality
    Used in Theorem 1 to bound false positive and false negative probabilities; it is valid but applies to true score distributions, not directly to pseudo-labels.
  • domain assumption Pseudo-label fidelity
    Assumes the top-k nodes by anomaly score approximate the true anomalies well enough that maximizing CSM on pseudo-labels improves true AUC; not proven and can fail for poor configurations.
  • domain assumption Known anomaly ratio
    Section 6.5.3 assumes the anomaly ratio is approximately known to set k, which is a form of prior information that may come from labels.
  • domain assumption Comparable loss scales
    Section 5.1.3 requires loss scales across configurations to be approximately equal so CSM values are comparable across hyperparameter settings.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Automated Self-Supervised Learning for Truly Unsupervised Graph Anomaly Detection." pith.science (2026). https://pith.science/paper/7YSDGOPU

@misc{pith2026250114694,
  author       = {Pith},
  title        = {Pith review of: Towards Automated Self-Supervised Learning for Truly Unsupervised Graph Anomaly Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7YSDGOPU}},
  note         = {Machine review of arXiv:2501.14694}
}
read the original abstract

Self-supervised learning (SSL) is an emerging paradigm that exploits supervisory signals generated from the data itself, and many recent studies have leveraged SSL to conduct graph anomaly detection. However, we empirically found that three important factors can substantially impact detection performance across datasets: 1) the specific SSL strategy employed; 2) the tuning of the strategy's hyperparameters; and 3) the allocation of combination weights when using multiple strategies. Most SSL-based graph anomaly detection methods circumvent these issues by arbitrarily or selectively (i.e., guided by label information) choosing SSL strategies, hyperparameter settings, and combination weights. While an arbitrary choice may lead to subpar performance, using label information in an unsupervised setting is label information leakage and leads to severe overestimation of a method's performance. Leakage has been criticized as "one of the top ten data mining mistakes", yet many recent studies on SSL-based graph anomaly detection have been using label information to select hyperparameters. To mitigate this issue, we propose to use an internal evaluation strategy (with theoretical analysis) to select hyperparameters in SSL for unsupervised anomaly detection. We perform extensive experiments using 10 recent SSL-based graph anomaly detection algorithms on various benchmark datasets, demonstrating both the prior issues with hyperparameter selection and the effectiveness of our proposed strategy.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

116 extracted references · 66 canonical work pages

  1. [1]

    Data mining and knowledge discovery 29:626--688

    Akoglu L, Tong H, Koutra D (2015) Graph based anomaly detection and description: a survey. Data mining and knowledge discovery 29:626--688

    2015

  2. [2]

    International Journal of Data Science and Analytics 14(2):113--126

    Bahri M, Salutari F, Putina A, et al (2022) Automl: state of the art with a focus on anomaly detection, challenges, and research directions. International Journal of Data Science and Analytics 14(2):113--126

    2022

  3. [3]

    Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 13(2):e1484

    Bischl B, Binder M, Lang M, et al (2023) Hyperparameter optimization: Foundations, algorithms, best practices, and open challenges. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery 13(2):e1484

    2023

  4. [4]

    In: Proceedings of the 2000 ACM SIGMOD international conference on Management of data, pp 93--104

    Breunig MM, Kriegel HP, Ng RT, et al (2000) Lof: identifying density-based local outliers. In: Proceedings of the 2000 ACM SIGMOD international conference on Management of data, pp 93--104

    2000

  5. [5]

    IEEE Transactions on Knowledge and Data Engineering

    Chen B, Zhang J, Zhang X, et al (2022) Gccad: Graph contrastive learning for anomaly detection. IEEE Transactions on Knowledge and Data Engineering

    2022

  6. [6]

    In: International conference on machine learning, PMLR, pp 1597--1607

    Chen T, Kornblith S, Norouzi M, et al (2020 a ) A simple framework for contrastive learning of visual representations. In: International conference on machine learning, PMLR, pp 1597--1607

    2020

  7. [7]

    In: Proceedings of the 29th ACM International Conference on Information & Knowledge Management, pp 1989--1992

    Chen Z, Liu B, Wang M, et al (2020 b ) Generative adversarial attributed network anomaly detection. In: Proceedings of the 29th ACM International Conference on Information & Knowledge Management, pp 1989--1992

    2020

  8. [8]

    arXiv preprint arXiv:180509501

    Cubuk ED, Zoph B, Mane D, et al (2018) Autoaugment: Learning augmentation policies from data. arXiv preprint arXiv:180509501

    2018

Show all 116 references
  1. [9]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, pp 702--703

    Cubuk ED, Zoph B, Shlens J, et al (2020) Randaugment: Practical automated data augmentation with a reduced search space. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, pp 702--703

    2020

  2. [10]

    In: Proceedings of the 2019 SIAM International Conference on Data Mining, SIAM, pp 594--602

    Ding K, Li J, Bhanushali R, et al (2019) Deep anomaly detection on attributed networks. In: Proceedings of the 2019 SIAM International Conference on Data Mining, SIAM, pp 594--602

    2019

  3. [11]

    In: Proceedings of the twenty-ninth international conference on international joint conferences on artificial intelligence, pp 1288--1294

    Ding K, Li J, Agarwal N, et al (2021) Inductive anomaly detection on attributed networks. In: Proceedings of the twenty-ninth international conference on international joint conferences on artificial intelligence, pp 1288--1294

    2021

  4. [12]

    Advances in Neural Information Processing Systems 35:9603--9616

    Ding X, Zhao L, Akoglu L (2022) Hyperparameter sensitivity in deep outlier detection: Analysis and a scalable hyper-ensemble solution. Advances in Neural Information Processing Systems 35:9603--9616

    2022

  5. [13]

    In: Proceedings of the AAAI Conference on Artificial Intelligence, pp 7459--7467

    Duan J, Wang S, Zhang P, et al (2023) Graph anomaly detection via multi-scale contrastive learning networks with augmented view. In: Proceedings of the AAAI Conference on Artificial Intelligence, pp 7459--7467

    2023

  6. [14]

    arXiv preprint arXiv:190512614

    Duan S, Matthey L, Saraiva A, et al (2019) Unsupervised model selection for variational disentangled representation learning. arXiv preprint arXiv:190512614

    2019

  7. [15]

    In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 5685--5689

    Fan H, Zhang F, Li Z (2020) Anomalydae: Dual autoencoder for anomaly detection on attributed networks. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, pp 5685--5689

    2020

  8. [16]

    arXiv preprint arXiv:190302428

    Fey M, Lenssen JE (2019) Fast graph representation learning with pytorch geometric. arXiv preprint arXiv:190302428

    2019

  9. [17]

    computers & security 28(1-2):18--28

    Garcia-Teodoro P, Diaz-Verdejo J, Maci \'a -Fern \'a ndez G, et al (2009) Anomaly-based network intrusion detection: Techniques, systems and challenges. computers & security 28(1-2):18--28

    2009

  10. [18]

    Goix N (2016) How to evaluate the quality of unsupervised anomaly detection algorithms? arXiv preprint arXiv:160701152

    2016

  11. [19]

    Advances in neural information processing systems 27

    Goodfellow I, Pouget-Abadie J, Mirza M, et al (2014) Generative adversarial nets. Advances in neural information processing systems 27

    2014

  12. [20]

    Radiology 143(1):29--36

    Hanley JA, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristic (roc) curve. Radiology 143(1):29--36

    1982

  13. [21]

    arXiv preprint arXiv:220109830

    Hassani K, Khasahmadi AH (2022) Learning graph augmentations to learn graph representations. arXiv preprint arXiv:220109830

    2022

  14. [22]

    In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XXV 16, Springer, pp 1--16

    Hataya R, Zdenek J, Yoshizoe K, et al (2020) Faster autoaugment: Learning augmentation strategies using backpropagation. In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XXV 16, Springer, pp 1--16

    2020

  15. [23]

    In: International conference on machine learning, PMLR, pp 2731--2741

    Ho D, Liang E, Chen X, et al (2019) Population based augmentation: Efficient learning of augmentation policy schedules. In: International conference on machine learning, PMLR, pp 2731--2741

    2019

  16. [24]

    In: 2020 IEEE international conference on data mining (ICDM), IEEE, pp 222--231

    Jiao Y, Xiong Y, Zhang J, et al (2020) Sub-graph contrast for scalable self-supervised graph representation learning. In: 2020 IEEE international conference on data mining (ICDM), IEEE, pp 222--231

    2020

  17. [25]

    In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pp 3122--3126

    Jin M, Liu Y, Zheng Y, et al (2021 a ) Anemone: Graph anomaly detection with multi-scale contrastive learning. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pp 3122--3126

    2021

  18. [26]

    arXiv preprint arXiv:210605470

    Jin W, Liu X, Zhao X, et al (2021 b ) Automated self-supervised learning for graphs. arXiv preprint arXiv:210605470

    2021

  19. [27]

    Journal of Global optimization 13:455--492

    Jones DR, Schonlau M, Welch WJ (1998) Efficient global optimization of expensive black-box functions. Journal of Global optimization 13:455--492

    1998

  20. [28]

    ACM Transactions on Knowledge Discovery from Data (TKDD) 6(4):1--21

    Kaufman S, Rosset S, Perlich C, et al (2012) Leakage in data mining: Formulation, detection, and avoidance. ACM Transactions on Knowledge Discovery from Data (TKDD) 6(4):1--21

    2012

  21. [29]

    IEEE Access

    Kim H, Lee BS, Shin WY, et al (2022) Graph anomaly detection with graph neural networks: Current status and challenges. IEEE Access

    2022

  22. [30]

    arXiv preprint arXiv:160902907

    Kipf TN, Welling M (2016) Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:160902907

    2016

  23. [31]

    Journal of the ACM (JACM) 46(5):604--632

    Kleinberg JM (1999) Authoritative sources in a hyperlinked environment. Journal of the ACM (JACM) 46(5):604--632

    1999

  24. [32]

    In: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pp 1269--1278

    Kumar S, Zhang X, Leskovec J (2019) Predicting dynamic embedding trajectory in temporal interaction networks. In: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, pp 1269--1278

    2019

  25. [33]

    In: Proceedings of the aaai conference on artificial intelligence, pp 16060--16062

    Lai KH, Zha D, Wang G, et al (2021) Tods: An automated time series outlier detection system. In: Proceedings of the aaai conference on artificial intelligence, pp 16060--16062

    2021

  26. [34]

    Advances in neural information processing systems 25

    Leskovec J, Mcauley J (2012) Learning to discover social circles in ego networks. Advances in neural information processing systems 25

    2012

  27. [35]

    In: IJCAI, pp 2152--2158

    Li J, Dani H, Hu X, et al (2017 a ) Radar: Residual analysis for anomaly detection in attributed networks. In: IJCAI, pp 2152--2158

    2017

  28. [36]

    Information Sciences 382:350--358

    Li Y, Shang Y, Yang Y (2017 b ) Clustering coefficients of large networks. Information Sciences 382:350--358

    2017

  29. [37]

    In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XXII 16, Springer, pp 580--595

    Li Y, Hu G, Wang Y, et al (2020 a ) Differentiable automatic data augmentation. In: Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XXII 16, Springer, pp 580--595

    2020

  30. [38]

    In: Companion Proceedings of the Web Conference 2020, pp 153--157

    Li Y, Zha D, Venugopal P, et al (2020 b ) Pyodds: An end-to-end outlier detection system with automated machine learning. In: Companion Proceedings of the Web Conference 2020, pp 153--157

    2020

  31. [39]

    IEEE Transactions on Neural Networks and Learning Systems 33(6):2365--2377

    Li Y, Chen Z, Zha D, et al (2021 a ) Automated anomaly detection via curiosity-guided search and self-imitation learning. IEEE Transactions on Neural Networks and Learning Systems 33(6):2365--2377

    2021

  32. [40]

    In: 2021 IEEE 37th International Conference on Data Engineering (ICDE), IEEE, pp 2117--2122

    Li Y, Chen Z, Zha D, et al (2021 b ) Autood: Neural architecture search for outlier detection. In: 2021 IEEE 37th International Conference on Data Engineering (ICDE), IEEE, pp 2117--2122

    2021

  33. [41]

    IEEE Transactions on Knowledge and Data Engineering 36(12):7839--7850

    Li Z, Liang S, Shi J, et al (2024 a ) Cross-domain graph level anomaly detection. IEEE Transactions on Knowledge and Data Engineering 36(12):7839--7850

    2024

  34. [42]

    In: Proceedings of the 2024 IEEE/ACM 46th International Conference on Software Engineering: Companion Proceedings, pp 306--307

    Li Z, Shi J, Van Leeuwen M (2024 b ) Graph neural networks based log anomaly detection and explanation. In: Proceedings of the 2024 IEEE/ACM 46th International Conference on Software Engineering: Companion Proceedings, pp 306--307

    2024

  35. [43]

    Advances in Neural Information Processing Systems 32

    Lim S, Kim I, Kim T, et al (2019) Fast autoaugment. Advances in Neural Information Processing Systems 32

    2019

  36. [44]

    In: international conference on machine learning, PMLR, pp 6127--6139

    Lin Z, Thekumparampil K, Fanti G, et al (2020) Infogan-cr and modelcentrality: Self-supervised model training and selection for disentangling gans. In: international conference on machine learning, PMLR, pp 6127--6139

    2020

  37. [45]

    In: 2022 IEEE International Conference on Data Mining (ICDM), IEEE, pp 259--268

    Liu F, Ma X, Wu J, et al (2022 a ) Dagad: Data augmentation for graph anomaly detection. In: 2022 IEEE International Conference on Data Mining (ICDM), IEEE, pp 259--268

    2022

  38. [46]

    Advances in Neural Information Processing Systems 35:27021--27035

    Liu K, Dou Y, Zhao Y, et al (2022 b ) Bond: Benchmarking unsupervised outlier node detection on static attributed graphs. Advances in Neural Information Processing Systems 35:27021--27035

    2022

  39. [47]

    IEEE transactions on neural networks and learning systems 33(6):2378--2392

    Liu Y, Li Z, Pan S, et al (2021) Anomaly detection on attributed networks via contrastive self-supervised learning. IEEE transactions on neural networks and learning systems 33(6):2378--2392

    2021

  40. [48]

    IEEE Transactions on Knowledge and Data Engineering 35(6):5879--5900

    Liu Y, Jin M, Pan S, et al (2022 c ) Graph self-supervised learning: A survey. IEEE Transactions on Knowledge and Data Engineering 35(6):5879--5900

    2022

  41. [49]

    arXiv preprint arXiv:230502496

    Liu Z, Cao C, Tao F, et al (2023) Revisiting graph contrastive learning for anomaly detection. arXiv preprint arXiv:230502496

    2023

  42. [50]

    arXiv preprint arXiv:220213248

    Luo Y, McThrow M, Au WY, et al (2022) Automated data augmentations for graph classification. arXiv preprint arXiv:220213248

    2022

  43. [51]

    ACM SIGKDD Explorations Newsletter 25(1)

    Ma MQ, Zhao Y, Zhang X, et al (2023) The need for unsupervised outlier model selection: A review and evaluation of internal evaluation strategies. ACM SIGKDD Explorations Newsletter 25(1)

    2023

  44. [52]

    In: Proceedings of the fifteenth ACM international conference on web search and data mining, pp 704--714

    Ma R, Pang G, Chen L, et al (2022) Deep graph-level anomaly detection by glocal knowledge distillation. In: Proceedings of the fifteenth ACM international conference on web search and data mining, pp 704--714

    2022

  45. [53]

    IEEE Transactions on Knowledge and Data Engineering

    Ma X, Wu J, Xue S, et al (2021) A comprehensive survey on graph anomaly detection with deep learning. IEEE Transactions on Knowledge and Data Engineering

    2021

  46. [54]

    In: Proceedings of the 27th international conference on scientific and statistical database management, pp 1--12

    Marques HO, Campello RJ, Zimek A, et al (2015) On the internal evaluation of unsupervised outlier detection. In: Proceedings of the 27th international conference on scientific and statistical database management, pp 1--12

    2015

  47. [55]

    ACM Transactions on Knowledge Discovery from Data (TKDD) 14(4):1--42

    Marques HO, Campello RJ, Sander J, et al (2020) Internal evaluation of unsupervised outlier detection. ACM Transactions on Knowledge Discovery from Data (TKDD) 14(4):1--42

    2020

  48. [56]

    Expert Systems with Applications p 122156

    Motie S, Raahemi B (2023) Financial fraud detection using graph neural networks: A systematic review. Expert Systems with Applications p 122156

    2023

  49. [57]

    Journal of Computer Science and Cybernetics 32(3):259--272

    Nguyen TT, Nguyen UQ, et al (2016) An evaluation method for unsupervised anomaly detection algorithms. Journal of Computer Science and Cybernetics 32(3):259--272

    2016

  50. [58]

    Academic press

    Nisbet R, Elder J, Miner GD (2009) Handbook of statistical analysis and data mining applications. Academic press

    2009

  51. [59]

    Advances in neural information processing systems 32

    Paszke A, Gross S, Massa F, et al (2019) Pytorch: An imperative style, high-performance deep learning library. Advances in neural information processing systems 32

    2019

  52. [60]

    In: IJCAI, pp 3513--3519

    Peng Z, Luo M, Li J, et al (2018) Anomalous: A joint modeling approach for anomaly detection on attributed networks. In: IJCAI, pp 3513--3519

    2018

  53. [61]

    In: Proceedings of the 2016 SIAM International Conference on Data Mining, SIAM, pp 207--215

    Perozzi B, Akoglu L (2016) Scalable anomaly ranking of attributed neighborhoods. In: Proceedings of the 2016 SIAM International Conference on Data Mining, SIAM, pp 207--215

    2016

  54. [62]

    In: 2022 IEEE 9th International Conference on Data Science and Advanced Analytics (DSAA), IEEE, pp 1--10

    Putina A, Bahri M, Salutari F, et al (2022) Autoad: an automated framework for unsupervised anomaly detectio. In: 2022 IEEE 9th International Conference on Data Science and Advanced Analytics (DSAA), IEEE, pp 1--10

    2022

  55. [63]

    Advances in Neural Information Processing Systems 36

    Qiao H, Pang G (2024) Truncated affinity maximization: One-class homophily modeling for graph anomaly detection. Advances in Neural Information Processing Systems 36

    2024

  56. [64]

    Advances in neural information processing systems 30

    Ratner AJ, Ehrenberg H, Hussain Z, et al (2017) Learning to compose domain-specific transformations for data augmentation. Advances in neural information processing systems 30

    2017

  57. [65]

    In: Proceedings of the 21th acm sigkdd international conference on knowledge discovery and data mining, pp 985--994

    Rayana S, Akoglu L (2015) Collective opinion spam detection: Bridging review networks and metadata. In: Proceedings of the 21th acm sigkdd international conference on knowledge discovery and data mining, pp 985--994

    2015

  58. [66]

    In: 2013 IEEE 13th international conference on data mining, IEEE, pp 647--656

    S \'a nchez PI, M \"u ller E, Laforet F, et al (2013) Statistical selection of congruent subspaces for mining attributed graphs. In: 2013 IEEE 13th international conference on data mining, IEEE, pp 647--656

    2013

  59. [67]

    AI magazine 29(3):93--93

    Sen P, Namata G, Bilgic M, et al (2008) Collective classification in network data. AI magazine 29(3):93--93

    2008

  60. [68]

    In: 2021 International Joint Conference on Neural Networks (IJCNN), IEEE, pp 1--7

    Sun J, Wang B, Wu B (2021) Automated graph representation learning for node classification. In: 2021 International Joint Conference on Neural Networks (IJCNN), IEEE, pp 1--7

    2021

  61. [69]

    Advances in Neural Information Processing Systems 34:15920--15933

    Suresh S, Li P, Hao C, et al (2021) Adversarial graph augmentation to improve graph contrastive learning. Advances in Neural Information Processing Systems 34:15920--15933

    2021

  62. [70]

    In: Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pp 990--998

    Tang J, Zhang J, Yao L, et al (2008) Arnetminer: extraction and mining of academic social networks. In: Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pp 990--998

    2008

  63. [71]

    In: Sixth international conference on data mining (ICDM'06), IEEE, pp 613--622

    Tong H, Faloutsos C, Pan JY (2006) Fast random walk with restart and its applications. In: Sixth international conference on data mining (ICDM'06), IEEE, pp 613--622

    2006

  64. [72]

    arXiv preprint arXiv:171010903

    Veli c kovi \'c P, Cucurull G, Casanova A, et al (2017) Graph attention networks. arXiv preprint arXiv:171010903

    2017

  65. [73]

    Advances in Neural Information Processing Systems 33:6514--6527

    Wenzel F, Snoek J, Tran D, et al (2020) Hyperparameter ensembles for robustness and uncertainty quantification. Advances in Neural Information Processing Systems 33:6514--6527

    2020

  66. [74]

    Advances in neural information processing systems 8

    Williams C, Rasmussen C (1995) Gaussian processes for regression. Advances in neural information processing systems 8

    1995

  67. [75]

    IEEE Transactions on Knowledge and Data Engineering

    Wu L, Lin H, Tan C, et al (2021) Self-supervised learning on graphs: Contrastive, generative, or predictive. IEEE Transactions on Knowledge and Data Engineering

    2021

  68. [76]

    IEEE Transactions on Pattern Analysis & Machine Intelligence 13(08):841--847

    Xie XL, Beni G (1991) A validity measure for fuzzy clustering. IEEE Transactions on Pattern Analysis & Machine Intelligence 13(08):841--847

    1991

  69. [77]

    IEEE transactions on pattern analysis and machine intelligence 45(2):2412--2429

    Xie Y, Xu Z, Zhang J, et al (2022) Self-supervised learning of graph neural networks: A unified review. IEEE transactions on pattern analysis and machine intelligence 45(2):2412--2429

    2022

  70. [78]

    In: Proceedings of the ACM Web Conference 2022, pp 2501--2510

    Xu W, Wu J, Liu Q, et al (2022 a ) Evidence-aware fake news detection with graph neural networks. In: Proceedings of the ACM Web Conference 2022, pp 2501--2510

    2022

  71. [79]

    In: 2019 IEEE International Conference on Big Data (Big Data), IEEE, pp 4201--4207

    Xu Z, Kakde D, Chaudhuri A (2019) Automatic hyperparameter tuning method for local outlier factor, with applications to anomaly detection. In: 2019 IEEE International Conference on Big Data (Big Data), IEEE, pp 4201--4207

    2019

  72. [80]

    In: Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, pp 444--457

    Xu Z, Huang X, Zhao Y, et al (2022 b ) Contrastive attributed network anomaly detection with data augmentation. In: Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, pp 444--457

    2022

  73. [81]

    Neurocomputing 415:295--316

    Yang L, Shami A (2020) On hyperparameter optimization of machine learning algorithms: Theory and practice. Neurocomputing 415:295--316

    2020

  74. [82]

    In: Proceedings of the AAAI conference on artificial intelligence, pp 8892--8900

    Yin Y, Wang Q, Huang S, et al (2022) Autogcl: Automated graph contrastive learning via learnable view generators. In: Proceedings of the AAAI conference on artificial intelligence, pp 8892--8900

    2022

  75. [83]

    In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Springer, pp 254--269

    Yoo J, Zhao Y, Zhao L, et al (2023) Dsv: An alignment validation loss for self-supervised outlier model selection. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Springer, pp 254--269

    2023

  76. [84]

    Advances in neural information processing systems 33:5812--5823

    You Y, Chen T, Sui Y, et al (2020) Graph contrastive learning with augmentations. Advances in neural information processing systems 33:5812--5823

    2020

  77. [85]

    In: International Conference on Machine Learning, PMLR, pp 12121--12132

    You Y, Chen T, Shen Y, et al (2021) Graph contrastive learning automated. In: International Conference on Machine Learning, PMLR, pp 12121--12132

    2021

  78. [86]

    In: 2021 IEEE International Conference on Big Data (Big Data), IEEE, pp 2691--2700

    Yuan X, Zhou N, Yu S, et al (2021) Higher-order structure based anomaly detection on attributed networks. In: 2021 IEEE International Conference on Big Data (Big Data), IEEE, pp 2691--2700

    2021

  79. [87]

    Advances in Neural Information Processing Systems 35:29350--29361

    Yue H, Zhang C, Zhang C, et al (2022) Label-invariant augmentation for semi-supervised graph classification. Advances in Neural Information Processing Systems 35:29350--29361

    2022

  80. [88]

    arXiv preprint arXiv:190704931

    Zeng H, Zhou H, Srivastava A, et al (2019) Graphsaint: Graph sampling based inductive learning method. arXiv preprint arXiv:190704931

    2019

  81. [89]

    In: Proceedings of the AAAI conference on Artificial Intelligence, pp 10824--10832

    Zeng J, Xie P (2021) Contrastive self-supervised learning for graph classification. In: Proceedings of the AAAI conference on Artificial Intelligence, pp 10824--10832

    2021

  82. [90]

    In: 2020 IEEE International Conference on Data Mining (ICDM), IEEE, pp 771--780

    Zha D, Lai KH, Wan M, et al (2020) Meta-aad: Active anomaly detection with deep reinforcement learning. In: 2020 IEEE International Conference on Data Mining (ICDM), IEEE, pp 771--780

    2020

  83. [91]

    arXiv preprint arXiv:220504816

    Zhang J, Wang S, Chen S (2022) Reconstruction enhanced multi-view contrastive learning for anomaly detection on attributed networks. arXiv preprint arXiv:220504816

    2022

  84. [92]

    arXiv preprint arXiv:230401565

    Zhang M, Qamar M, Kang T, et al (2023) A survey on graph diffusion models: Generative ai in science for molecule, protein and material. arXiv preprint arXiv:230401565

    2023

  85. [93]

    arXiv preprint arXiv:191211188

    Zhang X, Wang Q, Zhang J, et al (2019) Adversarial autoaugment. arXiv preprint arXiv:191211188

    2019

  86. [94]

    In: Proceedings of the aaai conference on artificial intelligence, pp 11015--11023

    Zhao T, Liu Y, Neves L, et al (2021 a ) Data augmentation for graph neural networks. In: Proceedings of the aaai conference on artificial intelligence, pp 11015--11023

    2021

  87. [95]

    In: Learning on Graphs Conference, PMLR, pp 32--1

    Zhao T, Tang X, Zhang D, et al (2022 a ) Autogda: Automated graph data augmentation for node classification. In: Learning on Graphs Conference, PMLR, pp 32--1

    2022

  88. [96]

    arXiv preprint arXiv:220811727

    Zhao Y, Akoglu L (2022) Towards unsupervised hpo for outlier detection. arXiv preprint arXiv:220811727

    2022

  89. [97]

    In: AutoML 2024 Methods Track

    Zhao Y, Akoglu L (2024) Hpod: Hyperparameter optimization for unsupervised outlier detection. In: AutoML 2024 Methods Track

    2024

  90. [98]

    In: Proceedings of the 2019 SIAM International Conference on Data Mining, SIAM, pp 585--593

    Zhao Y, Nasrullah Z, Hryniewicki MK, et al (2019) Lscp: Locally selective combination in parallel outlier ensembles. In: Proceedings of the 2019 SIAM International Conference on Data Mining, SIAM, pp 585--593

    2019

  91. [99]

    arXiv preprint arXiv:200910606

    Zhao Y, Rossi RA, Akoglu L (2020) Automating outlier detection via meta-learning. arXiv preprint arXiv:200910606

    2020

  92. [100]

    Advances in Neural Information Processing Systems 34:4489--4502

    Zhao Y, Rossi R, Akoglu L (2021 b ) Automatic unsupervised outlier model selection. Advances in Neural Information Processing Systems 34:4489--4502

    2021

  93. [101]

    In: 2022 IEEE International Conference on Data Mining (ICDM), IEEE, pp 773--782

    Zhao Y, Zhang S, Akoglu L (2022 b ) Toward unsupervised outlier model selection. In: 2022 IEEE International Conference on Data Mining (ICDM), IEEE, pp 773--782

    2022

  94. [102]

    IEEE Transactions on Knowledge and Data Engineering

    Zheng Y, Jin M, Liu Y, et al (2021) Generative and contrastive self-supervised learning for graph anomaly detection. IEEE Transactions on Knowledge and Data Engineering

    2021

  95. [103]

    In: Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining, pp 665--674

    Zhou C, Paffenroth RC (2017) Anomaly detection with robust deep autoencoders. In: Proceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining, pp 665--674

    2017

  96. [104]

    write newline

    " write newline " cite write " FUNCTION editor.postfix editor num.names #1 > "( )" "( )" if FUNCTION editor.trans.postfix editor num.names #1 > "( )" "( )" if FUNCTION trans.postfix translator num.names #1 > "( )" "( )" if FUNCTION authors.editors.reflist.apa5 'field := 'dot :...

  97. [105]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter doi edition editor eid howpublished institution journal key keywords month note number organization pages publisher school series title type url volume year eprint archive archivePrefix primaryClass adsurl adsnote version label INTEGERS o...

  98. [106]

    write newline

    " write newline "" before.all 'output.state := FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap duplicate "7" = swap duplicate "8" = swap "9" = or or or or or or...

  99. [107]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter doi edition editor eid eprint howpublished institution journal key keywords month note number organization pages publisher school series title type url volume year archivePrefix primaryClass adsurl adsnote version label extra.labe...

  100. [108]

    write newline

    " write newline "" before.all 'output.state := FUNCTION add.period duplicate empty 'skip "." * add.blank if FUNCTION if.digit duplicate "0" = swap duplicate "1" = swap duplicate "2" = swap duplicate "3" = swap duplicate "4" = swap duplicate "5" = swap duplicate "6" = swap dupl...

  101. [109]

    write newline

    " write newline "" before.all 'output.state := FUNCTION output.doi doi empty skip "doi:" doi * "" * output if FUNCTION format.archive archivePrefix empty "" archivePrefix ":" * if FUNCTION format.primaryClass primaryClass empty "" " [" primaryClass * "] " * if FUNCTION format....

  102. [110]

    write newline

    " write newline "" before.all 'output.state := FUNCTION string.to.integer 't := t text.length 'k := #1 'char.num := t char.num #1 substring 's := s is.num s "." = or char.num k = not and char.num #1 + 'char.num := while char.num #1 - 'char.num := t #1 char.num substring FUNCTI...

  103. [111]

    write newline

    " write newline "" before.all 'output.state := FUNCTION string.to.integer 't := t text.length 'k := #1 'char.num := t char.num #1 substring 's := s is.num s "." = or char.num k = not and char.num #1 + 'char.num := while char.num #1 - 'char.num := t #1 char.num substring FUNCTI...

  104. [112]

    , " * write output.state after.block = add.period write newline

    ENTRY address archive author booktitle chapter edition editor eprint howpublished institution journal key keywords month note number organization pages publisher school series title type url doi volume year archivePrefix primaryClass eid adsurl adsnote version label INTEGERS o...

  105. [113]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  106. [114]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  107. [115]

    Available from:

    ENTRY address assignee author booktitle chapter cartographer day edition editor howpublished institution inventor journal key keywords month note number organization pages part publisher school series title type volume word year eprint doi url lastchecked updated archive archi...

  108. [116]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.