Pith. sign in

REVIEW 3 major objections 4 minor 45 references

Know Your Account: Double Graph Inference-based Account De-anonymization on Ethereum

T0 review · 3 major / 4 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 Reading each Ethereum address through both its static transaction network and that network's evolution over time, DBG4ETH identifies account types at F1 above 97 percent, beating single-view models by up to 40 points and prior methods by…

desk verdict A well-engineered double-graph account classifier whose SOTA claim currently rests on an under-specified calibration split; worth refereeing but not yet citable. read the letter →

arxiv 2411.18875 v2 pith:YLCZG5MH submitted 2024-11-28 cs.SI

Shuyi Miao , Wangjie Qiu , Hongwei Zheng , Qinnan Zhang , Xiaofan Tu , Xunan Liu , Yang Liu , Jin Dong
show 1 more author
Zhiming Zheng
This is my paper · ORCID
classification cs.SI
keywords Ethereumaccountde-anonymizationidentificationgraphneuralnetworksdynamicconfidencecalibrationtransactionnetworkblockchainregulation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that the type of an anonymous Ethereum account — exchange, wallet, miner, or phisher/hacker — can be inferred from public transaction history alone, with F1 scores above 97 percent on all four categories, by learning two complementary views of the same data: a global static graph of all interactions and a local dynamic graph of how those interactions evolve across time slices. The proposed pipeline, DBG4ETH, samples labeled subgraphs, encodes the static view with a hierarchical attention network plus contrastive learning and the dynamic view with a GCN–GRU over ten time slices, then fuses the two branches' predictions through an adaptive calibration module that weights six confidence-calibration methods by how much each reduces expected calibration error. On the reported dataset, the fused model beats each graph branch alone by 3.75 to 40.52 F1 points and outperforms the compared baselines by 5.23 to 12.91 points, while keeping near-optimal performance with only 20 to 30 percent of the training data and on newly emerging account types. If the results hold, regulators and exchanges could screen large numbers of addresses and act on calibrated confidence scores rather than on raw labels.

What carries the argument

The load-bearing mechanism is the pairing of two complementary graph encoders with an ECE-weighted calibration fusion. The Global Static Graph (GSG) keeps the full transaction neighborhood as one static subgraph, so the hierarchical GAT encoder sees long-horizon structure that time-slicing would discard; the Local Dynamics Graph (LDG) cuts the same subgraph into ten time slices and runs a GCN–GRU with Diffpool, so the model also sees short-horizon evolution that a static snapshot hides. The two branches' predictions are merged by adaptive confidence calibration: each of six calibrators (three parametric, three non-parametric) recalibrates each branch, and equation (25) assigns weight $\alpha_i = \Delta ECE_i / \sum_{j=1}^6 \Delta ECE_j$, the share of total expected-calibration-error reduction each method contributes, so better-calibrating methods dominate the fused score. A LightGBM classifier then turns the two weighted calibration results into an account type.

What would settle it

Fix the equation (25) calibration weights on a validation fold strictly disjoint from the test set, retrain LightGBM on training labels only, and remeasure F1 on the four account types; if the frozen-weight F1 falls toward the best baseline's level instead of the reported 99.51/97.19/97.56/98.42, the adaptive calibration was borrowing test information. A companion check is corrupting 10 percent of the public labels and watching whether precision collapses, which would implicate label quality rather than the architecture.

Watch

Extended reading notes

Core claim

DBG4ETH casts Ethereum account de-anonymization as a subgraph classification problem. Around each labeled account it samples a two-hop subgraph of top-K transaction partners, decorates every node with 15 engineered behavioral features (send and receive counts and values, time-interval statistics, gas fees, contract-call counts), and reads the subgraph twice: the global static branch, a hierarchical graph attention network regularized by contrastive learning with adaptive augmentation, compresses the whole history into a long-term behavioral fingerprint, while the local dynamic branch splits the same history into ten time slices and runs a GCN–GRU with differentiable pooling to capture how behavior drifts. The two branches output scores that are rescaled, passed through six calibration methods (temperature scaling, beta and logistic calibration, histogram binning, isotonic regression, and Bayesian binning into quantiles), and merged by weights proportional to each method's measured reduction in expected calibration error, after which a LightGBM classifier maps the two weighted score vectors to an identity. The paper reports F1 scores of 99.51 (Exchange), 97.19 (ICO-Wallet), 97.56 (Mining), and 98.42 (Phish/Hack), beating each single-graph branch by 3.75–40.52 points and prior identity-inference baselines by 5.23–12.91 points.

Load-bearing premise

The headline F1 numbers depend on an unstated data split: the calibration weights in equation (25) and the LightGBM classifier must see only training or validation data, never the test accounts, but the paper does not report that split, the number of runs, or the random seeds, so if test predictions contributed to the ECE weights the reported gains would overstate generalization.

Editorial extensions

If this is right

  • Two views beat one: because similar static graphs can hide very different histories and similar histories can arise from different static structures, fusing the global-static and local-dynamic encoders is what lifts F1 by 3.75 to 40.52 points over either branch alone, so single-view graph models leave recoverable signal unused.
  • Confidence becomes actionable: the calibration module converts raw scores into reliability-weighted probabilities, so downstream decisions such as freezing an account or flagging a wallet for review can be gated by a confidence threshold rather than by a hard label.
  • Label scarcity is less crippling than feared: with as few as 56 labeled mining accounts, and with only 20 to 30 percent of the training set for bridge and DeFi accounts, DBG4ETH reaches near-optimal F1, and it transfers to newly emerging account types without architectural changes.
  • Hand-built features still carry weight: adding the 15-dimensional behavioral features lifts a plain GCN's F1 by 16.73 points on the same data, so engineered features and graph architecture are complements, not substitutes.

Reading between the lines

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

  • Read as a deployment blueprint, DBG4ETH is a triage tool rather than a proof of identity: with calibrated probabilities in hand, the natural workflow is to investigate high-confidence anomalous accounts first, but the paper does not simulate that workflow, measure its cost, or propose an operating threshold.
  • The strongest untested assumption sits inside equation (25): the ECE-reduction weights must be computed on data the classifier has not seen. A clean follow-up would freeze those weights on a validation fold and report test F1 with seeds and run counts, which would protect the claimed 5–13 point edge over baselines.
  • The method's ceiling is set by the public label clouds it trains on, so a stress test the paper does not run is label-noise injection: flipping or dropping a fraction of labels and measuring how quickly the four-category F1 degrades would separate architectural gains from label-quality gains.
  • The static/dynamic complementarity is generic, so the recipe should transfer to any timestamped ledger such as Bitcoin or other smart-contract chains; re-running DBG4ETH with the same 15 features on another chain's labeled accounts would test that transfer cheaply.
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 / 4 minor

Summary. The paper proposes DBG4ETH, a double-graph framework for Ethereum account de-anonymization. It constructs a global static graph (GSG) and a local dynamic graph (LDG) from Ethereum transaction data, encodes them with GNNs plus contrastive learning, and then feeds the weighted calibrated probabilities from six calibration methods into a LightGBM classifier. Experiments report state-of-the-art F1 scores for exchange, ICO-wallet, mining, and phish/hack accounts, and additional results for bridge and DeFi accounts. The main claimed contributions are the combination of static/dynamic and global/local perspectives, an adaptive confidence calibration module, and strong empirical performance under label scarcity.

Significance. If the reported results are reliable, the method is a meaningful advance: it explicitly addresses both static/dynamic and global/local transaction perspectives, tackles confidence calibration, which is often ignored, and ships source code. However, the empirical claims are currently not verifiable because the evaluation protocol omits a documented data split, random seeds, and repeated runs, and because the adaptive calibration weights may be computed using test-set information. These are load-bearing gaps: the central SOTA claim depends on the reported F1 scores being unbiased estimates of generalization.

major comments (3)
  1. [Section V-A4, Table III] The experimental section reports no train/validation/test split, no random seeds, no repeated runs, and no error bars. With per-class positive samples as small as 56 for Mining (Table II), the single-point F1 values (e.g., 97.56 for Mining) could be high-variance noise. The paper must specify the split ratio (or cross-validation scheme), report the number of runs with seeds, and provide error bars or at least the variance across runs to support the SOTA claim.
  2. [Section IV-C3, Eqs. (24)-(25)] The adaptive calibration weights are computed from ECE reductions (ΔECE), but the manuscript never states whether the ECE used in Eq. (25) is evaluated on a held-out calibration set disjoint from the test set used for the final F1. If test examples contribute to the ECE reductions, the weighted probabilities P encode test-set information and the reported F1 is not a valid generalization estimate. The paper must state the calibration split and its size, and explain how overlapping subgraphs (where the same address can be the center of one subgraph and a neighbor of another) are prevented from leaking across the split.
  3. [Section IV-C3 and V-B2] Eq. (25) permits negative α_i when ΔECE is negative, which the paper acknowledges in V-B2 ('there are cases where parametric calibration methods receive negative weights'). However, the paper does not describe any clamping, renormalization, or other procedure that keeps the weighted prediction P in Eq. (24) within [0,1]. Without such a procedure, P is not a calibrated probability and may produce out-of-range inputs for the LightGBM classifier. This issue should be resolved and the resulting effect on F1 reported.
minor comments (4)
  1. [Table III] The 'Improve.' row reports improvements as percentages but does not name the baseline to which DBG4ETH is compared; the reader has to infer it from the table. Please specify the baseline explicitly (e.g., the best-performing competitor for each column).
  2. [Table V] The header 'Pricision' is a typo; it should be 'Precision'.
  3. [Figures 4-7, 9] Several figures contain garbled text (likely PDF extraction artifacts, e.g., '/uni000000...' strings). These should be cleaned in the camera-ready version so that axis labels and legends are readable.
  4. [Section III-B1, Eq. (2)] Equation (2) is notationally ambiguous: the symbol K is used both as the number of sampled neighbors and as an index set. Please clarify the notation, e.g., by defining the neighborhood set and the sampling operator explicitly.

Circularity Check

0 steps flagged · score 1.0 of 10

No demonstrated circularity: the pipeline is standard supervised learning on external labels; the underspecified calibration/evaluation split is a reproducibility concern, not an equation-level circular step.

full rationale

The central identification claim rests on a supervised pipeline: 15-dimensional transaction features, GSG/LDG graph encoders, an adaptive calibration module (Eq. 24-25), and LightGBM classification, trained and evaluated against ground-truth labels obtained from Etherscan/XLabelCloud. These labels are external to the model, so the core classification result is not defined in terms of its own output. The only potentially self-referential component is the adaptive weight α_i = ΔECE_i / ΣΔECE_i (Eq. 25), since ΔECE is computed by comparing calibrated predictions with ground-truth labels. If the same labeled examples were used both to compute these ECE reductions and to report the final F1, then the headline numbers would be training-set scores rather than generalization estimates. However, the paper never states that the calibration set is the same as the test set; it simply omits the train/calibration/test split, the number of runs, and random seeds (Sections IV-C3 and V-A4). Under the hard rule that circularity must be exhibited as an explicit reduction rather than inferred from missing documentation, this omission is a serious reproducibility and correctness risk, but not a demonstrated circular step. The self-citations to TEGDetector [13] and the GNN survey [14], which share coauthor Hongwei Zheng, are used only as a baseline and as background, respectively, and are not load-bearing for the proposed method's validity. No uniqueness theorem is imported from the authors' prior work, no ansatz is smuggled in via self-citation, and no known result is merely renamed. Therefore, no circular step can be quoted and exhibited from the paper's own equations; the honest finding is no significant circularity, with the calibration-split ambiguity flagged as an external validity threat rather than a circularity.

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

No new physical entities are introduced; the double graph is an architectural construct. The free parameters are standard hand-set hyperparameters plus the ECE-derived calibration weights. The main axioms are data-quality and sampling assumptions inherited from the Ethereum label sources.

free parameters (5)
  • neighbor sampling K and hops h = K=2000, h=2
    Set in Section V-A4; controls subgraph size and the account representation; hand-chosen.
  • LDG time slices and pooling configuration = T=10, pooling layers=2, cluster ratios N*0.1 and 1
    Set in Section V-A4; sensitivity is tested only for pooling layers, not for T.
  • contrastive augmentation probabilities = Pe=(0.3, 0.4), Pf=(0.1, 0.0)
    Set in Section V-A4; sensitivity analysis shows stable behavior for small values.
  • learning rate = chosen from {0.1, 0.05, 0.01, 0.005, 0.001}
    Tuned with Adam in Section V-A4; the final selected value is not reported.
  • adaptive calibration weights = alpha_i per branch, computed from Delta ECE
    Eq. 25; these weights are fitted to the calibration data and drive the claimed confidence improvement.
assumptions (4)
  • domain assumption Ethereum labels from Etherscan/XLabelCloud correctly identify account categories
    Section V-A1 uses these platforms as ground truth; label noise or mislabeling would directly change the reported F1.
  • domain assumption Transaction data from Xblock for 2015-2024 are complete and correctly parsed
    All graphs and features are derived from this source; missing blocks or parse errors would bias the behavioral features.
  • domain assumption A 2-hop top-K subgraph preserves sufficient identity signal
    The method classifies the center account from this sampled subgraph in Section III-B1; if distant or low-value transactions carry identity, the model cannot see them.
  • domain assumption ECE reduction is a valid proxy for better calibrated predictions
    The adaptive weighting in Eqs. 24-25 is driven entirely by ECE change; if ECE is insensitive, the claimed calibration benefit is not established.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Know Your Account: Double Graph Inference-based Account De-anonymization on Ethereum." pith.science (2026). https://pith.science/paper/YLCZG5MH

@misc{pith2026241118875,
  author       = {Pith},
  title        = {Pith review of: Know Your Account: Double Graph Inference-based Account De-anonymization on Ethereum},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YLCZG5MH}},
  note         = {Machine review of arXiv:2411.18875}
}
read the original abstract

The scaled Web 3.0 digital economy, represented by decentralized finance (DeFi), has sparked increasing interest in the past few years, which usually relies on blockchain for token transfer and diverse transaction logic. However, illegal behaviors, such as financial fraud, hacker attacks, and money laundering, are rampant in the blockchain ecosystem and seriously threaten its integrity and security. In this paper, we propose a novel double graph-based Ethereum account de-anonymization inference method, dubbed DBG4ETH, which aims to capture the behavioral patterns of accounts comprehensively and has more robust analytical and judgment capabilities for current complex and continuously generated transaction behaviors. Specifically, we first construct a global static graph to build complex interactions between the various account nodes for all transaction data. Then, we also construct a local dynamic graph to learn about the gradual evolution of transactions over different periods. Different graphs focus on information from different perspectives, and features of global and local, static and dynamic transaction graphs are available through DBG4ETH. In addition, we propose an adaptive confidence calibration method to predict the results by feeding the calibrated weighted prediction values into the classifier. Experimental results show that DBG4ETH achieves state-of-the-art results in the account identification task, improving the F1-score by at least 3.75% and up to 40.52% compared to processing each graph type individually and outperforming similar account identity inference methods by 5.23% to 12.91%.

Figures

Figures reproduced from arXiv: 2411.18875 by the authors.

Figure 1
Figure 1. Details about the challenge (i): (a) and (b) have similar [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The overview of our DBG4ETH. determined by three measures: degree centrality [29], eigen￾vector centrality [29], and PageRank centrality [29]. (2) Node￾attribute-level augmentation: node attribute augmentation by randomly masking a fraction of dimensions with zeros in node features. We note the labels obtained for the generated graph as the labels of the original graph as well. Subgraph contrast learning on the GSG … view at source ↗
Figure 3
Figure 3. Adaptive calibration method for two types of prediction [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 5
Figure 5. Figure 5: Scatter plot of account category features. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Different types of accounts use different calibration methods to obtain weight proportion data adaptively. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: ROC curves for different classifiers selected on four different accounts. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: The impact of training set size on model performance. [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Effect of hyperparameters in GSG and LDG encoder [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 34 canonical work pages

  1. [1]

    Blockchain challenges and opportunities: A survey,

    Z. Zheng, S. Xie, H.-N. Dai, X. Chen, and H. Wang, “Blockchain challenges and opportunities: A survey,” International journal of web and grid services, vol. 14, no. 4, pp. 352–375, 2018

    2018

  2. [2]

    Applications of blockchain technology beyond cryptocurrency,

    M. H. Miraz and M. Ali, “Applications of blockchain technology beyond cryptocurrency,” CoRR, vol. abs/1801.03528, 2018

    arXiv 2018

  3. [3]

    Cryptocurrency transaction network em- bedding from static and dynamic perspectives: An overview,

    Y . Zhou, X. Luo, and M. Zhou, “Cryptocurrency transaction network em- bedding from static and dynamic perspectives: An overview,”IEEE/CAA Journal of Automatica Sinica, vol. 10, no. 5, pp. 1105–1121, 2023

    work page 2023

  4. [4]

    Sequence-based target coin prediction for cryptocurrency pump-and-dump,

    S. Hu, Z. Zhang, S. Lu, B. He, and Z. Li, “Sequence-based target coin prediction for cryptocurrency pump-and-dump,” Proceedings of the ACM on Management of Data, vol. 1, no. 1, pp. 1–19, 2023

    work page 2023

  5. [5]

    When digital economy meets web3. 0: Applications and challenges,

    C. Chen, L. Zhang, Y . Li, T. Liao, S. Zhao, Z. Zheng, H. Huang, and J. Wu, “When digital economy meets web3. 0: Applications and challenges,” IEEE Open Journal of the Computer Society, vol. 3, pp. 233–245, 2022

    work page 2022

  6. [6]

    Artificial intelligence for web 3.0: A comprehensive survey,

    M. Shen, Z. Tan, D. Niyato, Y . Liu, J. Kang, Z. Xiong, L. Zhu, W. Wang, and X. Shen, “Artificial intelligence for web 3.0: A comprehensive survey,” ACM Computing Surveys, 2023

    work page 2023

  7. [7]

    Zipzap: Efficient training of language models for large-scale fraud detection on blockchain,

    S. Hu, T. Huang, K.-H. Chow, W. Wei, Y . Wu, and L. Liu, “Zipzap: Efficient training of language models for large-scale fraud detection on blockchain,” in Proceedings of the ACM on Web Conference 2024, 2024, pp. 2807–2816

    work page 2024

  8. [8]

    Denseflow: Spot- ting cryptocurrency money laundering in ethereum transaction graphs,

    D. Lin, J. Wu, Y . Yu, Q. Fu, Z. Zheng, and C. Yang, “Denseflow: Spot- ting cryptocurrency money laundering in ethereum transaction graphs,” in Proceedings of the ACM on Web Conference 2024, 2024, pp. 4429– 4438

    work page 2024

Show all 45 references
  1. [9]

    Identity inference on blockchain using graph neural network,

    J. Shen, J. Zhou, Y . Xie, S. Yu, and Q. Xuan, “Identity inference on blockchain using graph neural network,” in Blockchain and Trustworthy Systems: Third International Conference, BlockSys 2021, Guangzhou, China, August 5–6, 2021, Revised Selected Papers 3. Springer, 2021, pp. 3–17

    2021

  2. [10]

    Behavior- aware account de-anonymization on ethereum interaction graph,

    J. Zhou, C. Hu, J. Chi, J. Wu, M. Shen, and Q. Xuan, “Behavior- aware account de-anonymization on ethereum interaction graph,” IEEE Transactions on Information Forensics and Security, vol. 17, pp. 3433– 3448, 2022

    2022

  3. [11]

    Riskprop: Account risk rating on ethereum via de-anonymous score and network propagation,

    D. Lin, J. Wu, Q. Fu, Z. Zheng, and T. Chen, “Riskprop: Account risk rating on ethereum via de-anonymous score and network propagation,” arXiv preprint arXiv:2301.00354, 2023

    2023 arXiv

  4. [12]

    Who is who on ethereum? account labeling using heterophilic graph convolutional net- work,

    D. Lin, J. Wu, T. Huang, K. Lin, and Z. Zheng, “Who is who on ethereum? account labeling using heterophilic graph convolutional net- work,” IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 54, no. 3, pp. 1541–1553, 2024

    2024

  5. [13]

    Tegdetector: A phishing detector that knows evolving transaction behaviors,

    H. Zheng, M. Ma, H. Ma, J. Chen, H. Xiong, and Z. Yang, “Tegdetector: A phishing detector that knows evolving transaction behaviors,” IEEE Transactions on Computational Social Systems, pp. 1–13, 2023

    2023

  6. [14]

    Graph neural networks: Taxonomy, advances, and trends,

    Y . Zhou, H. Zheng, X. Huang, S. Hao, D. Li, and J. Zhao, “Graph neural networks: Taxonomy, advances, and trends,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 13, no. 1, pp. 1–54, 2022

    2022

  7. [15]

    Both homophily and heterophily matter: Bi-path aware graph neural network for ethereum account classification,

    H. Yang, J. Fang, J. Wu, and Z. Zheng, “Both homophily and heterophily matter: Bi-path aware graph neural network for ethereum account classification,” IEEE J. Emerg. Sel. Topics Circuits Syst., vol. 13, no. 3, pp. 829–840, 2023

    2023

  8. [16]

    Bert4eth: a pre- trained transformer for ethereum fraud detection,

    S. Hu, Z. Zhang, B. Luo, S. Lu, B. He, and L. Liu, “Bert4eth: a pre- trained transformer for ethereum fraud detection,” in Proceedings of the ACM Web Conference 2023, 2023, pp. 2189–2197

    2023

  9. [17]

    Semi-supervised classification with graph convolutional networks,

    T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” CoRR, vol. abs/1609.02907, 2016

    2016 arXiv

  10. [18]

    Graph attention networks,

    P. Velickovic, G. Cucurull, A. Casanova, A. Romero, P. Lio, Y . Bengio et al., “Graph attention networks,” stat, vol. 1050, no. 20, pp. 10–48 550, 2017

    2017

  11. [19]

    How powerful are graph neural networks?

    K. Xu, W. Hu, J. Leskovec, and S. Jegelka, “How powerful are graph neural networks?” CoRR, vol. abs/1810.00826, 2018

    2018 arXiv

  12. [20]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems, vol. 30, 2017

    2017

  13. [21]

    Deepwalk: online learning of social representations,

    B. Perozzi, R. Al-Rfou, and S. Skiena, “Deepwalk: online learning of social representations,” in Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ser. KDD ’14. New York, NY , USA: Association for Computing Machinery, 2014, p. ...

    2014

  14. [22]

    Efficient estimation of word representations in vector space,

    T. Mikolov, K. Chen, G. Corrado, and J. Dean, “Efficient estimation of word representations in vector space,” in 1st International Conference on Learning Representations, ICLR 2013, Scottsdale, Arizona, USA, May 2-4, 2013, Workshop Track Proceedings, Y . Bengio and Y . LeCun, ...

    2013

  15. [23]

    node2vec: Scalable feature learning for networks,

    A. Grover and J. Leskovec, “node2vec: Scalable feature learning for networks,” in Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining, 2016, pp. 855– 864

    2016

  16. [24]

    Detecting phishing scams on ethereum based on transaction records,

    Q. Yuan, B. Huang, J. Zhang, J. Wu, H. Zhang, and X. Zhang, “Detecting phishing scams on ethereum based on transaction records,” in2020 IEEE international symposium on circuits and systems (ISCAS). IEEE, 2020, pp. 1–5

    2020

  17. [25]

    Trans2vec: learning transaction embedding via items and frequent itemsets,

    D. Nguyen, T. D. Nguyen, W. Luo, and S. Venkatesh, “Trans2vec: learning transaction embedding via items and frequent itemsets,” in Advances in Knowledge Discovery and Data Mining: 22nd Pacific-Asia Conference, PAKDD 2018, Melbourne, VIC, Australia, June 3-6, 2018, Proceedings,...

    2018

  18. [26]

    Hierarchical attention networks for document classification,

    Z. Yang, D. Yang, C. Dyer, X. He, A. Smola, and E. Hovy, “Hierarchical attention networks for document classification,” in Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: human language technologies, 2016, pp. ...

    2016

  19. [27]

    Graph con- trastive learning with adaptive augmentation,

    Y . Zhu, Y . Xu, F. Yu, Q. Liu, S. Wu, and L. Wang, “Graph con- trastive learning with adaptive augmentation,” in Proceedings of the Web Conference 2021, 2021, pp. 2069–2080

    2021

  20. [28]

    Cs231n convolutional neural networks for visual recognition,

    F.-F. Li, J. Johnson, and S. Yeung, “Cs231n convolutional neural networks for visual recognition,” https://cs231n.github.io/ neural-networks-1/#actfun, 2024, accessed: 2024-09-28

    2024

  21. [29]

    M ¨uller, J

    B. M ¨uller, J. Reinhardt, and M. T. Strickland, Neural networks: an introduction. Springer Science & Business Media, 2012

    2012

  22. [30]

    Learning phrase representations using RNN encoder-decoder for statistical machine translation,

    K. Cho, B. van Merrienboer, C ¸ . G ¨ulc ¸ehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y . Bengio, “Learning phrase representations using RNN encoder-decoder for statistical machine translation,” inProceedings of the 2014 Conference on Empirical Methods in Natural Language ...

    2014

  23. [31]

    Hierarchical graph representation learning with differentiable pooling,

    Z. Ying, J. You, C. Morris, X. Ren, W. L. Hamilton, and J. Leskovec, “Hierarchical graph representation learning with differentiable pooling,” in Advances in Neural Information Processing Systems 31: Annual Conference on Neural Information Processing Systems 2018, NeurIPS 2018...

    2018

  24. [32]

    Learnable structural semantic readout for graph classification,

    D. Lee, S. Kim, S. Lee, C. Park, and H. Yu, “Learnable structural semantic readout for graph classification,” in 2021 IEEE International Conference on Data Mining (ICDM). IEEE, 2021, pp. 1180–1185

    2021

  25. [33]

    On calibration of modern neural networks,

    C. Guo, G. Pleiss, Y . Sun, and K. Q. Weinberger, “On calibration of modern neural networks,” in International conference on machine learning. PMLR, 2017, pp. 1321–1330

    2017

  26. [34]

    Non-parametric calibration of probabilistic regression,

    H. Song, M. Kull, and P. A. Flach, “Non-parametric calibration of probabilistic regression,” CoRR, vol. abs/1806.07690, 2018

    2018 arXiv

  27. [35]

    Obtaining calibrated probability estimates from decision trees and naive bayesian classifiers,

    B. Zadrozny and C. Elkan, “Obtaining calibrated probability estimates from decision trees and naive bayesian classifiers,” in Icml, vol. 1, 2001, pp. 609–616

    2001

  28. [36]

    Transforming classifier scores into accurate multiclass probability estimates,

    ——, “Transforming classifier scores into accurate multiclass probability estimates,” in Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining, 2002, pp. 694– 699. 13

    2002

  29. [37]

    Obtaining well calibrated probabilities using bayesian binning,

    M. P. Naeini, G. Cooper, and M. Hauskrecht, “Obtaining well calibrated probabilities using bayesian binning,” in Proceedings of the AAAI conference on artificial intelligence, vol. 29, no. 1, 2015

    2015

  30. [38]

    Lightgbm: A highly efficient gradient boosting decision tree,

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T. Liu, “Lightgbm: A highly efficient gradient boosting decision tree,” in Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, ...

    2017

  31. [39]

    Inductive representation learning on large graphs,

    W. Hamilton, Z. Ying, and J. Leskovec, “Inductive representation learning on large graphs,” in Advances in Neural Information Processing Systems, I. Guyon, U. V . Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, Eds., vol. 30. Curran Associates, Inc....

    2017

  32. [40]

    Personalized embedding propagation: Combining neural networks on graphs with personalized pagerank,

    J. Klicpera, A. Bojchevski, and S. G ¨unnemann, “Personalized embedding propagation: Combining neural networks on graphs with personalized pagerank,” CoRR, vol. abs/1810.05997, 2018. [Online]. Available: http://arxiv.org/abs/1810.05997

    2018 arXiv

  33. [41]

    Graph inductive biases in transformers without message passing,

    L. Ma, C. Lin, D. Lim, A. Romero-Soriano, P. K. Dokania, M. Coates, P. Torr, and S.-N. Lim, “Graph inductive biases in transformers without message passing,” in International Conference on Machine Learning. PMLR, 2023, pp. 23 321–23 337

    2023

  34. [42]

    Tsgn: Transaction subgraph networks assisting phishing detection in ethereum,

    J. Wang, P. Chen, X. Xu, J. Wu, M. Shen, Q. Xuan, and X. Yang, “Tsgn: Transaction subgraph networks assisting phishing detection in ethereum,” arXiv preprint arXiv:2208.12938, 2022

    2022 arXiv

  35. [43]

    Random forests,

    L. Breiman, “Random forests,” Machine learning, vol. 45, pp. 5–32, 2001

    2001

  36. [44]

    Experiments with a new boosting algo- rithm,

    Y . Freund and R. E. Schapire, “Experiments with a new boosting algo- rithm,” in Machine Learning, Proceedings of the Thirteenth International Conference (ICML ’96), Bari, Italy, July 3-6, 1996, L. Saitta, Ed. Morgan Kaufmann, 1996, pp. 148–156

    1996

  37. [45]

    Xgboost: A scalable tree boosting system,

    T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785–794. 14

    2016

Pith tools

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