Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

FreCT: Frequency-augmented Convolutional Transformer for Robust Time Series Anomaly Detection

T0 review · 3 major / 7 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read The paper claims that an unsupervised, patch-based contrastive transformer that aligns two views in both time and frequency domains outperforms reconstruction-based and other transformer baselines on four multivariate anomaly detection…

desk verdict A plausible anomaly-detection architecture whose printed training loss doesn't match its anomaly score — fix that and the small F1 gains might hold; right now they're not reproducible from the text. read the letter →

arxiv 2505.00941 v2 pith:W3ZPESA6 submitted 2025-05-02 cs.LG

Wenxin Zhang , Ding Xu , Guangzhen Yao , Xiaojian Lin , Renxiang Guan , Chengze Du , Renda Han , Xi Xuan
show 1 more author
Cuicui Luo
This is my paper · ORCID
classification cs.LG
keywords timeseriesanomalydetectioncontrastivelearningfrequencydomainFouriertransformTransformerconvolutionKLdivergenceunsupervised
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

FreCT is an unsupervised method for multivariate time series anomaly detection that learns normal patterns from two contrastive views of the same sequence—one across patches and one within patches—and measures how consistently those views agree in both the time and frequency domains. The central claim is that replacing reconstruction error with a stop-gradient KL-divergence consistency objective, augmented by Fourier-domain consistency, protects the learned normal pattern from contamination by anomalous segments and captures information that time-domain analysis alone misses. On the MSL, SMAP, SWaT, and PSM datasets, FreCT reports F1 values of 95.26, 96.51, 96.20, and 97.55 percent, exceeding the best baseline by 1.01, 0.1, 1.98, and 1.83 points respectively. A sympathetic reader would care because the method needs no anomaly labels and no negative samples, and because the frequency-domain consistency term is a separable, testable addition to existing contrastive approaches.

What carries the argument

The load-bearing mechanism is a pair of contrastive views generated by patching the time series: an inter-patch view (attention across patches) and an intra-patch view (attention within patches), encoded with shared weights and refined by multi-scale parallel convolution with kernel sizes 1, 3, and 5. Consistency is enforced by a symmetric KL-divergence loss with stop-gradient in the time domain and by an absolute-error loss between FFT magnitudes in the frequency domain; the two losses are combined with a mixing weight α, and the same combined score is used at inference. Patching supplies local semantic structure, the shared-weight Transformer supplies long-range dependency modeling, and stop-gradient prevents the consistency objective from being satisfied by a trivial constant embedding.

What would settle it

Train FreCT on a multivariate series where anomalous segments occupy more than half the training window; if the consistency-based anomaly score then tracks the anomalies instead of the normal pattern because the stop-gradient KL has aligned to anomalous structure, the rarity assumption that grounds the method is falsified.

Watch

Extended reading notes

Core claim

The paper claims that anomaly detection in multivariate time series can be reframed as a consistency problem between two patch-level views of the same input, rather than a reconstruction problem. The inter-patch view treats each patch as a unit and models dependencies between patches; the intra-patch view models dependencies among elements inside a patch. Both views are encoded by a shared-weight Transformer with parallel multi-scale convolution, then compared with a symmetric KL divergence that applies stop-gradient on one side, so the model learns to align the normal pattern without collapsing to a trivial solution. In parallel, the encoded views are transformed with the fast Fourier transform, and consistency is measured as absolute difference in frequency components, adding information that time-domain auto-correlation obscures. The paper reports that this combined consistency score outperforms eleven baselines on four public datasets, and its ablations attribute the gain to the normalization, convolution, and frequency modules.

Load-bearing premise

Anomalies are rare enough that the encoded embeddings of normal points from the two patch views stay consistent, so the stop-gradient KL objective learns the normal pattern rather than being pulled toward anomalous segments.

Editorial extensions

If this is right

  • Unsupervised time series anomaly detection can be improved by measuring cross-view consistency instead of reconstruction error, without anomaly labels or negative samples.
  • The frequency-domain consistency term is separable: ablating it lowers average F1 by about 0.97 points across the four datasets, so Fourier information carries signal beyond the time-domain view.
  • The stop-gradient symmetric KL loss is not just regularization; ablating it on one or both views degrades performance, so robustness depends specifically on that mechanism.
  • Because the same consistency score is used for training and inference, the method yields a point-wise anomaly score that can be thresholded directly, with no separate reconstruction head or post-hoc scoring procedure.
  • Reported training time is competitive with or shorter than several transformer baselines on most datasets, making the dual-view consistency objective a practical alternative to association-discrepancy and dual-autoencoder designs.

Reading between the lines

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

  • The paper does not test this, but the same dual-view consistency objective with Fourier augmentation could be applied to online or streaming anomaly detection, where resistance to contamination from anomalous segments matters most.
  • A natural extension is irregularly sampled or missing-data series: the FFT step would require interpolation, and the consistency loss would then measure how well both views agree on reconstructed spectra, a regime the paper does not address.
  • The rarity assumption suggests a stress test: on datasets with long or frequent anomaly segments in the training window, the consistency score should be monitored for drift, because the KL objective may begin aligning to anomaly patterns if anomalies are no longer rare.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 7 minor

Summary. FreCT proposes an unsupervised contrastive-learning architecture for multivariate time series anomaly detection. It generates inter-patch and intra-patch views via patching, encodes them with a shared Transformer plus multi-scale convolution, adds a Fourier-frequency consistency branch, and trains with a stop-gradient KL divergence loss in the time domain and an absolute-error loss in the frequency domain. At test time, anomaly scores are computed as the alpha-weighted sum of the time-domain and frequency-domain consistency terms. Experiments on MSL, SMAP, SWaT, and PSM report F1 scores above all eleven baselines, with ablations and sensitivity studies.

Significance. The architectural idea—contrastive consistency between patch-level views, augmented by frequency-domain alignment with stop-gradient KL—is interesting and could be a useful contribution to unsupervised time series anomaly detection. However, the empirical claims are not yet firmly supported: the training objective and the test-time score are inconsistent as written, the threshold selection protocol for converting scores to binary labels is unspecified, and the reported tables contain no variance information. If the training-loss issue is corrected and the evaluation protocol clarified, the method could be a solid contribution; as it stands, the reported state-of-the-art result cannot be verified from the text.

major comments (3)
  1. [IV-F, Eq. (20), and IV-G, Eq. (22)] The time-domain training loss is L_tim = (L_{Z'_P} - L_{Z'_N})/len, where L_{Z'_P} and L_{Z'_N} are defined in Eqs. (18)-(19). The time-domain anomaly score in Eq. (22) is Score_tim(X) = Σ KL(Z'_P, Ω(Z'_N)) + KL(Ω(Z'_P), Z'_N), which is not equal to either L_{Z'_P} or L_{Z'_N}; it mixes the first term of Eq. (18) with the second term of Eq. (19). Consequently, gradient descent minimizes the difference between two directional consistency losses, while the test-time detection score is a different functional of the same representations. A model can reduce L_tim by balancing the two directional terms while leaving Score_tim large for normal data, which would directly inflate the anomaly scores at inference. If Eq. (20) is a typo for L_tim = (L_{Z'_P} + L_{Z'_N})/len, the stated objective is not the objective that produced Table II; either way, the central performance claim is not reproducible from the text as written.
  2. [IV-G, Eq. (25)] The binary detection threshold ρ is introduced as a hyperparameter, but the manuscript never states how ρ is chosen. In particular, it is not specified whether ρ is selected per dataset, on a held-out validation set, or by optimizing F1 on the test labels. Since the reported F1 metric is entirely determined by this threshold, the absolute numbers in Table II cannot be independently checked without this protocol. Additionally, the model operates on windows (patched sequences) while Eq. (25) labels individual points; the paper does not describe how window-level scores are converted to point-level predictions (e.g., center-point assignment, sliding-window majority, or all-points-within-window).
  3. [Table II and Section V-C] All results are single-run with no standard deviations, seeds, or significance tests. Several claimed gains are very small; for example, the F1 improvement over AnomalyTrans on SMAP is stated as 0.1 percentage point (96.51 vs. 96.41). Without repeated runs and a measure of variance, the claim that FreCT outperforms existing methods is not statistically supported. The same limitation applies to the ablation tables (Tables III-V), where differences of fractions of a point are interpreted as demonstrating the utility of individual components.
minor comments (7)
  1. [II-A] The sentence 'We select eleven algorithms as baselines, including two traditional anomaly detection methods ... and seven deep learning approaches' is inconsistent with the enumeration that follows: LOF and DAGMM are traditional, while VAE, OmniAnomaly, TranAD, AnomalyTrans, DCFF-MTAD, MAUT, ATF-UAD, BTAD, and GIN are nine deep learning methods, not seven.
  2. [IV-B and Figure 2a] Equation (1) uses Var(·) in the denominator, but Figure 2a and the surrounding text refer to standard deviation; the formula and notation should be made consistent.
  3. [IV-C, Eqs. (3)-(5)] The shapes of the upsampled attention weights and value matrices are not made explicit enough to verify the tensor contraction in Eq. (5); please provide index notation or a shape table.
  4. [IV-E, Eq. (15)] The Fourier transform is written as an integral over infinite time for a finite-length discrete sequence; use the DFT definition appropriate to the patch length and clarify whether the transform is applied along the time axis or the patch axis.
  5. [Tables III and IV] Several entries appear with missing decimal points (e.g., '9996' should presumably be '99.96', and '9268' should be '92.68'), making the tables difficult to read.
  6. [V-B and Figure 4] The per-dataset hyperparameter configurations are reported in the text, but the notation in Figure 4 (e.g., '[15]' and '[135]') is not harmonized with the patch-size notation used elsewhere; please align the notation.
  7. [V-G and Figure 7] The training-time comparison would be more informative with standard deviations and with a clear statement of the hardware and software used for all compared methods.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FreCT is an externally benchmarked consistency-based anomaly detector; the training-loss/inference-score mismatch is a correctness concern, not a circular reduction.

full rationale

The paper's central claim is empirical: FreCT is compared against eleven external baselines on four public datasets, and its F1 numbers in Table II are independent of the method's own definitions. The anomaly score in Eq. 24 is a weighted combination of the time- and frequency-domain consistency losses, which is a standard design choice for reconstruction- and consistency-based anomaly detection rather than a circular derivation; the score is not fitted to test labels in any quoted procedure, and the threshold rho in Eq. 25 is described as a hyperparameter with no stated test-label fitting. There are no load-bearing self-citations: the cited prior works (DCdetector, Anomaly Transformer, SimSiam, BYOL, etc.) are external, and no uniqueness theorem or prior result by the same authors is invoked to force the architecture or objective. A potential internal inconsistency exists between Eq. 20, which defines the time-domain training loss as a difference of two KL terms, and Eq. 22, which evaluates a one-directional KL combination at inference; if Eq. 20 is literal, the model is not trained to minimize the exact score it evaluates. This is a reproducibility and correctness issue, but it is not circularity, because the reported performance is still validated against independent labeled benchmarks. The paper is self-contained with respect to its empirical evaluation, so the circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central contribution rests on one empirical objective and several hand-set constants. alpha, rho, and per-dataset hyperparameters are not reported or are chosen without a validation protocol; the domain assumptions about anomaly rarity and frequency consistency are asserted rather than derived. No new physical or conceptual entities are introduced.

free parameters (3)
  • alpha = not reported
    Eq. 21 and Eq. 24 use the same alpha to blend time and frequency losses; the value is never given, so the reported scores depend on an undisclosed manual choice.
  • detection threshold rho = not reported
    Eq. 25 binarizes the anomaly score; the selection procedure is not described, so rho may be tuned on test labels.
  • per-dataset architecture hyperparameters = patch and hidden sizes vary by dataset; window sizes incomplete
    Section V-B lists per-dataset settings chosen from sensitivity experiments; without a validation protocol these are fitted to each benchmark.
assumptions (4)
  • domain assumption Anomalies are rare and normal points from different contrastive views should be consistent in latent space
    Section IV-F states this incentive; if violated, the consistency score will not separate anomalies.
  • domain assumption Frequency-domain consistency beyond time-domain dependencies helps detect anomalies
    Section IV-E assumes Fourier information adds discriminative signal; no falsifiable theoretical derivation is given.
  • domain assumption Stop-gradient prevents collapse without negative samples
    Section IV-F cites SimSiam; the paper relies on this empirical result for training stability.
  • standard math FFT and KL divergence behave as standard implementations
    Equations 15-19 assume standard FFT and KL; reasonable but unverified for the unusual tensor shapes in Eqs. 4-9.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FreCT: Frequency-augmented Convolutional Transformer for Robust Time Series Anomaly Detection." pith.science (2026). https://pith.science/paper/W3ZPESA6

@misc{pith2026250500941,
  author       = {Pith},
  title        = {Pith review of: FreCT: Frequency-augmented Convolutional Transformer for Robust Time Series Anomaly Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W3ZPESA6}},
  note         = {Machine review of arXiv:2505.00941}
}
read the original abstract

Time series anomaly detection is critical for system monitoring and risk identification, across various domains, such as finance and healthcare. However, for most reconstruction-based approaches, detecting anomalies remains a challenge due to the complexity of sequential patterns in time series data. On the one hand, reconstruction-based techniques are susceptible to computational deviation stemming from anomalies, which can lead to impure representations of normal sequence patterns. On the other hand, they often focus on the time-domain dependencies of time series, while ignoring the alignment of frequency information beyond the time domain. To address these challenges, we propose a novel Frequency-augmented Convolutional Transformer (FreCT). FreCT utilizes patch operations to generate contrastive views and employs an improved Transformer architecture integrated with a convolution module to capture long-term dependencies while preserving local topology information. The introduced frequency analysis based on Fourier transformation could enhance the model's ability to capture crucial characteristics beyond the time domain. To protect the training quality from anomalies and improve the robustness, FreCT deploys stop-gradient Kullback-Leibler (KL) divergence and absolute error to optimize consistency information in both time and frequency domains. Extensive experiments on four public datasets demonstrate that FreCT outperforms existing methods in identifying anomalies.

Figures

Figures reproduced from arXiv: 2505.00941 by the authors.

Figure 1
Figure 1. The overall framework of FreCT. First, FreCT leverages the preprocessing module to normalize the time series and generate patches. Then, FreCT [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. The sequence-level preprocessing, including sequence-level normal [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The rationality validation experiments of the loss function. [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: The sensitivity experimental results of patch size. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: The sensitivity experimental results of window size. [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: The F1 score metric of sensitivity experiments on the number of layers, the embedding size, and the number of heads. [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: The efficiency observation of FreCT compared with several baselines. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Butter: Frequency Consistency and Hierarchical Fusion for Autonomous Driving Object Detection

    cs.CV 2025-07 reject novelty 3.0 of 10

    A YOLO-family detector named Butter claims state-of-the-art efficiency on KITTI, BDD100K, and Cityscapes, but the paper's loss equations, parameter counts, and baseline comparisons contain contradictions that undermin...

Reference graph

Works this paper leans on

70 extracted references · 70 canonical work pages · cited by 1 Pith paper

  1. [1]

    Correlation-aware spatial–temporal graph learning for multivariate time-series anomaly detection,

    Y . Zheng, H. Y . Koh, M. Jin, L. Chi, K. T. Phan, S. Pan, Y .-P. P. Chen, and W. Xiang, “Correlation-aware spatial–temporal graph learning for multivariate time-series anomaly detection,”IEEE transactions on neural networks and learning systems , pp. 1–15, 2023

    work page 2023

  2. [2]

    Mag: A novel approach for effective anomaly detection in spacecraft telemetry data,

    B. Yu, Y . Yu, J. Xu, G. Xiang, and Z. Yang, “Mag: A novel approach for effective anomaly detection in spacecraft telemetry data,” IEEE transactions on industrial informatics , vol. 20, no. 3, pp. 3891 – 3899, 2024

    work page 2024

  3. [3]

    Disen- tangled anomaly detection for multivariate time series,

    X. Jie, X. Zhou, C. Su, Z. Zhou, Y . Yuan, J. Bu, and H. Wang, “Disen- tangled anomaly detection for multivariate time series,” in Proceedings of the ACM Web Conference , 2024, pp. 931–934

    work page 2024

  4. [4]

    Intricate anomaly detection in time series by caculating shift of non-gaussianity,

    M. Liu, D. Yang, F. Ahmad, and F. R. Beyette, “Intricate anomaly detection in time series by caculating shift of non-gaussianity,” in Proceedings of the IEEE SOUTHEASTCON , 2024, pp. 386–391

    work page 2024

  5. [5]

    Hybrid anomaly detection in time series by combining kalman filters and machine learning models,

    A. Puder, M. Zink, L. Seidel, and E. Sax, “Hybrid anomaly detection in time series by combining kalman filters and machine learning models,” Sensors, vol. 24, no. 9, 2024

    work page 2024

  6. [6]

    Units: A universal time series analysis framework powered by self-supervised representation learning,

    Z. Liang, C. Liang, Z. Liang, H. Wang, and B. Zheng, “Units: A universal time series analysis framework powered by self-supervised representation learning,” in Proceedings of the ACM SIGMOD interna- tional conference on management of data , 2024, pp. 480 – 483

    work page 2024

  7. [7]

    Unsupervised deep learning for iot time series,

    Y . Liu, Y . Zhou, K. Yang, and X. Wang, “Unsupervised deep learning for iot time series,” IEEE internet of things journal , vol. 10, no. 16, pp. 14 285 – 14 306, 2023

    work page 2023

  8. [8]

    Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges,

    G. Li and J. J. Jung, “Deep learning for anomaly detection in multivariate time series: Approaches, applications, and challenges,” Information fusion, vol. 91, pp. 93 – 102, 2023

    work page 2023

Show all 70 references
  1. [9]

    Time series diffusion in the frequency domain,

    J. Crabbe, N. Huynh, J. Stanczuk, and M. van der Schaar, “Time series diffusion in the frequency domain,” vol. 235, 2024, pp. 9407 – 9438

    2024

  2. [10]

    Time and frequency-domain feature fusion network for multivariate time series classification,

    T. Lei, J. Li, and K. Yang, “Time and frequency-domain feature fusion network for multivariate time series classification,” Expert Systems with Applications, vol. 252, 2024

    2024

  3. [11]

    Fits: Modeling time series with 10k parameters,

    Z. Xu, A. Zeng, and Q. Xu, “Fits: Modeling time series with 10k parameters,” in Proceedings of the 12th international conference on learning representations, 2023

    2023

  4. [12]

    Koopa: Learning non-stationary time series dynamics with koopman predictors,

    Y . Liu, C. Li, J. Wang, and M. Long, “Koopa: Learning non-stationary time series dynamics with koopman predictors,” in Proceedings of the 37th conference on neural information processing systems, vol. 36, 2023

    2023

  5. [13]

    Anomaly detection in time series: a comprehensive evaluation,

    S. Schmidl, P. Wenig, and T. Papenbrock, “Anomaly detection in time series: a comprehensive evaluation,” in Proceedings of the VLDB Endowment, vol. 15, no. 9, 2022, pp. 1779 – 1797

    2022

  6. [14]

    Robust monitoring conditional volatility change for time series based on support vector regression,

    M. H. Yoon, C. K. Kim, and S. Lee, “Robust monitoring conditional volatility change for time series based on support vector regression,” Communications in statistics-simulation and computation , 2024

    2024

  7. [15]

    Statistical process monitoring for vector autore- gressive time series based on location-scale cusum method,

    S. Lee and S. Lee, “Statistical process monitoring for vector autore- gressive time series based on location-scale cusum method,” Quality engineering, vol. 35, no. 3, pp. 493–518, 2023

    2023

  8. [16]

    The empty-nest power user management based on data mining technology,

    J. Li, J. Yang, H. Cai, C. Jiang, Q. Jiang, Y . Xie, Z. Lu, L. Li, and G. Sun, “The empty-nest power user management based on data mining technology,” Sensors, vol. 23, no. 5, 2023

    2023

  9. [17]

    Water flow modeling and forecast in a water branch of mexico city through arima and transfer function models for anomaly detection,

    D. Barrientos-Torres, E. A. Martinez-Rios, S. A. Navarro-Tuch, J. L. Pablos-Hach, and R. Bustamante-Bello, “Water flow modeling and forecast in a water branch of mexico city through arima and transfer function models for anomaly detection,” Water, vol. 15, no. 15, 2023

    2023

  10. [18]

    Early neurological deterioration detection with a transformer convolutional auto-encoder model,

    J. Yang, X. Nie, L. Wang, C. Huang, and L. Liu, “Early neurological deterioration detection with a transformer convolutional auto-encoder model,” Applied soft computing , vol. 151, 2024

    2024

  11. [19]

    A novel multi algorithm approach to identify network anomalies in the iot using fog computing and a model to distinguish between iot and non-iot devices,

    R. J. J. Alzahrani and A. Alzahrani, “A novel multi algorithm approach to identify network anomalies in the iot using fog computing and a model to distinguish between iot and non-iot devices,” Journal of sensor and actuator networks, vol. 12, no. 2, 2023

    2023

  12. [20]

    Clustering- based granular representation of time series with application to collective anomaly detection,

    W. Shi, D. Karastoyanova, Y . Ma, Y . Huang, and G. Zhang, “Clustering- based granular representation of time series with application to collective anomaly detection,” IEEE transactions on instrumentation and measure- ment, vol. 72, 2023

    2023

  13. [21]

    Machine learning-assisted improved anomaly detection for structural health monitoring,

    S. Samudra, M. Barbosh, and A. Sadhu, “Machine learning-assisted improved anomaly detection for structural health monitoring,” Sensors, vol. 23, no. 7, 2023

    2023

  14. [22]

    Hybrid amplitude ordinal partition networks for ecg morphology discrimination: an application to pvc recognition,

    Z. Cai, C. Ma, J. Li, and C. Liu, “Hybrid amplitude ordinal partition networks for ecg morphology discrimination: an application to pvc recognition,” IEEE transactions on instrumentation and measurement , vol. 73, 2024

    2024

  15. [23]

    Label-free multivariate time series anomaly detection,

    Q. Zhou, S. He, H. Liu, J. Chen, and W. Meng, “Label-free multivariate time series anomaly detection,” IEEE transactions on knowledge and data engineering, vol. 36, no. 7, pp. 3166–3179, 2024

    2024

  16. [24]

    Noncontact respiratory anomaly detection using infrared light- wave sensing,

    M. Z. Islam, B. Martin, C. Gotcher, T. Martinez, J. F. O’Hara, and S. Ekin, “Noncontact respiratory anomaly detection using infrared light- wave sensing,” IEEE transactions on human-machine systems , vol. 54, no. 3, pp. 292–303, 2024

    2024

  17. [25]

    Unsu- pervised constrained discord detection in iot-based online crane moni- toring,

    A. Mukherjee, M. Sasidharan, M. Herrera, and A. K. Parlikad, “Unsu- pervised constrained discord detection in iot-based online crane moni- toring,” Advanced engineering informatics , vol. 60, 2024

    2024

  18. [26]

    Time series anomaly detection with a transformer residual autoencoder-decoder,

    S. Wang, Y . Wang, and W. Li, “Time series anomaly detection with a transformer residual autoencoder-decoder,” in Proceedings of 30th International Conference on Neural Information Processing, vol. 14450, 2024, pp. 512–524

    2024

  19. [27]

    Unsupervised outlier detection for time- series data of indoor air quality using lstm autoencoder with ensemble method,

    J. Park, Y . Seo, and J. Cho, “Unsupervised outlier detection for time- series data of indoor air quality using lstm autoencoder with ensemble method,” Journal of big data , vol. 10, no. 1, 2023

    2023

  20. [28]

    Unsupervised detecting anomalies in multivariate time series by robust convolutional lstm encoder-decoder,

    T. Le, H. C. Vu, A. Ponchet-Durupt, N. Boudaoud, Z. Cherfi-Boulanger, and T. Nguyen-Trang, “Unsupervised detecting anomalies in multivariate time series by robust convolutional lstm encoder-decoder,” Neurocom- puting, vol. 592, 2024

    2024

  21. [29]

    Duogat: dual time-oriented graph attention networks for accurate, efficient and explainable anomaly de- tection on time-series,

    J. Lee, B. Park, and D.-K. Chae, “Duogat: dual time-oriented graph attention networks for accurate, efficient and explainable anomaly de- tection on time-series,” in Proceedings of the 32nd ACM international conference on information and knowledge management, 2023, pp. 1188– 1197

    2023

  22. [30]

    Spatial-temporal transformer with error-restricted variance estimation for time series anomaly detec- tion,

    Y . Feng, W. Zhang, H. Sun, and W. Jiang, “Spatial-temporal transformer with error-restricted variance estimation for time series anomaly detec- tion,” in Proceedings of the 28th Pacific-Asia Conference on Knowledge Discovery and Data Mining , vol. 14645, 2024, pp. 3–14

    2024

  23. [31]

    Dcdetector: dual attention contrastive representation learning for time series anomaly detection,

    Y . Yang, C. Zhang, T. Zhou, Q. Wen, and L. Sun, “Dcdetector: dual attention contrastive representation learning for time series anomaly detection,” in Proceedings of the 29th ACM SIGKDD conference on knowledge discovery and data mining , 2023, pp. 3033–3045

    2023

  24. [32]

    Anomaly transformer: time series anomaly detection with association discrepancy,

    J. Xu, H. Wu, J. Wang, and M. Long, “Anomaly transformer: time series anomaly detection with association discrepancy,” in Proceedings of 10th international conference on learning representations , 2022

    2022

  25. [33]

    Anomaly detection in time series with robust variational quasi- recurrent autoencoders,

    T. Kieu, B. Yang, C. Guo, R.-G. Cirstea, Y . Zhao, Y . Song, and C. S. Jensen, “Anomaly detection in time series with robust variational quasi- recurrent autoencoders,” in Proceedings of international conference on data engineering, vol. 2022, 2022, pp. 1342 – 1354

    2022

  26. [34]

    Gat-dns: Dns multivariate time series prediction model based on graph attention network,

    X. Lu, X. Zhang, and P. Lio, “Gat-dns: Dns multivariate time series prediction model based on graph attention network,” in Proceedings of the 32nd world wide web conference , 2023, pp. 127–131

    2023

  27. [35]

    From anomaly detection to classification with graph attention and transformer for multivariate time series,

    C. Wang and G. Liu, “From anomaly detection to classification with graph attention and transformer for multivariate time series,” Advanced Engineering Informatics, vol. 60, p. 102357, 2024

    2024

  28. [36]

    Decoupled contrastive learning for long-tailed distribution,

    X. Chen, Y . Zhou, L. Wang, D. Wu, W. Zhang, B. Li, and W. Wang, “Decoupled contrastive learning for long-tailed distribution,” in Pattern Recognition and Computer Vision , vol. 14433, 2024, pp. 3–15

    2024

  29. [37]

    Multi-view contrastive learning network for rec- ommendation,

    X. Bu and R. Ma, “Multi-view contrastive learning network for rec- ommendation,” in Pattern Recognition and Computer Vision , Q. Liu, H. Wang, Z. Ma, W. Zheng, H. Zha, X. Chen, L. Wang, and R. Ji, Eds., 2024, pp. 319–330

    2024

  30. [38]

    How does simsiam avoid collapse without negative samples? a unified understanding with self-supervised contrastive learning,

    C. Zhang, K. Zhang, C. Zhang, T. X. Pham, C. D. Yoo, and I. S. Kweon, “How does simsiam avoid collapse without negative samples? a unified understanding with self-supervised contrastive learning,” in Proceedings of the 10th international conference on learning representations , 2022

    2022

  31. [39]

    Representation uncertainty in self-supervised learning as variational inference,

    H. Nakamura, M. Okada, and T. Taniguchi, “Representation uncertainty in self-supervised learning as variational inference,” in Proceedings of the IEEE International Conference on Computer Vision , 2023, pp. 16 438–16 447

    2023

  32. [40]

    The edge of orthogonality: A simple view of what makes byol tick,

    P. H. Richemond, A. Tam, Y . Tang, F. Strub, B. Piot, and F. Hill, “The edge of orthogonality: A simple view of what makes byol tick,” in Proceedings of the 40th international conference on machine learning , vol. 202, 2023, pp. 29 063 – 29 081

    2023

  33. [41]

    Dual temperature helps contrastive learning without many negative samples: towards understanding and simplifying moco,

    C. Zhang, K. Zhang, T. X. Pham, A. Niu, Z. Qiao, C. D. Yoo, and I. S. Kweon, “Dual temperature helps contrastive learning without many negative samples: towards understanding and simplifying moco,” in Proceedings of the IEEE computer society conference on computer vision and p...

    2022

  34. [42]

    Fast-moco: Boost momentum- based contrastive learning with combinatorial patches,

    Y . Ci, C. Lin, L. Bai, and W. Ouyang, “Fast-moco: Boost momentum- based contrastive learning with combinatorial patches,” in Proceedings of the 17th European Conference on Computer Vision, vol. 13686, 2022, pp. 290 – 306

    2022

  35. [43]

    Implicit contrastive representation learning with guided stop-gradient,

    B. Lee and S. Lee, “Implicit contrastive representation learning with guided stop-gradient,” in Proceedings of the 37th conference on neural information processing systems , vol. 36, 2023

    2023

  36. [44]

    Recdcl: Dual contrastive learning for recommendation,

    D. Zhang, Y . Geng, W. Gong, Z. Qi, Z. Chen, X. Tang, Y . Shan, Y . Dong, and J. Tang, “Recdcl: Dual contrastive learning for recommendation,” in Proceedings of the ACM Web Conference , 2024, pp. 3655 – 3666

    2024

  37. [46]

    An efficient alternating algorithm for the l-norm cross- gradient joint inversion of gravity and magnetic data using the 2-d fast fourier transform,

    S. Vatankhah, S. Liu, R. A. Renaut, X. Hu, J. D. Hogue, and M. Gharloghi, “An efficient alternating algorithm for the l-norm cross- gradient joint inversion of gravity and magnetic data using the 2-d fast fourier transform,” IEEE transactions on geoscience and remote sensing, ...

    2022

  38. [47]

    Mining graph-fourier transform time series for anomaly detection of internet traffic at core and metro networks,

    M. Herrera, Y . Proselkov, M. Perez-Hernandez, and A. K. Parlikad, “Mining graph-fourier transform time series for anomaly detection of internet traffic at core and metro networks,” IEEE access , vol. 9, pp. 8997–9011, 2021

    2021

  39. [48]

    Cross-dimension attentive feature fusion network for unsupervised time-series anomaly detection,

    R. Wang, Y . Zhou, G. Luo, P. Chen, and D. Peng, “Cross-dimension attentive feature fusion network for unsupervised time-series anomaly detection,” CMES-computer modeling in engineering & sciences , vol. 139, no. 3, pp. 3011–3027, 2024

    2024

  40. [49]

    An adversarial time- frequency reconstruction network for unsupervised anomaly detection,

    J. Fan, Z. Wang, H. Wu, D. Sun, J. Wu, and X. Lu, “An adversarial time- frequency reconstruction network for unsupervised anomaly detection,” Neural networks, vol. 168, pp. 44–56, 2023

    2023

  41. [50]

    Diagnostic spatio-temporal transformer with faithful encoding,

    J. Labaien, T. Ide, P.-Y . Chen, E. Zugasti, and X. De Carlos, “Diagnostic spatio-temporal transformer with faithful encoding,” Knowledge-based systems, vol. 274, 2023

    2023

  42. [51]

    Itransformer: Inverted transformers are effective for time series fore- casting,

    Y . Liu, T. Hu, H. Zhang, H. Wu, S. Wang, L. Ma, and M. Long, “Itransformer: Inverted transformers are effective for time series fore- casting,” in Proceedings of the 25th international conference on learning representations, 2024

    2024

  43. [52]

    Non-stationary transformers: Exploring the stationarity in time series forecasting,

    Y . Liu, H. Wu, J. Wang, and M. Long, “Non-stationary transformers: Exploring the stationarity in time series forecasting,” in Proceedings of the 36th Conference on Neural Information Processing Systems, vol. 35, 2022

    2022

  44. [53]

    Incorporating convolution designs into visual transformers,

    K. Yuan, S. Guo, Z. Liu, A. Zhou, F. Yu, and W. Wu, “Incorporating convolution designs into visual transformers,” in Proceedings of the IEEE/CVF international conference on computer vision , 2021, pp. 579– 588

    2021

  45. [54]

    Dynamic unary convolution in transformers,

    H. Duan, Y . Long, S. Wang, H. Zhang, C. G. Willcocks, and L. Shao, “Dynamic unary convolution in transformers,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 11, pp. 12 747– 12 759, 2023

    2023

  46. [55]

    Conv2former: A simple transformer-style convnet for visual recognition,

    Q. Hou, C.-Z. Lu, M.-M. Cheng, and J. Feng, “Conv2former: A simple transformer-style convnet for visual recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence , no. 01, pp. 1–10, 2024

    2024

  47. [56]

    Fredf: Learning to forecast in frequency domain,

    H. Wang, L. Pan, Z. Chen, D. Yang, S. Zhang, Y . Yang, X. Liu, H. Li, and D. Tao, “Fredf: Learning to forecast in frequency domain,” in Proceedings of the 25th international conference on learning represen- tations, 2024

    2024

  48. [57]

    Exploring simple siamese representation learning,

    X. Chen and K. He, “Exploring simple siamese representation learning,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition , 2021, pp. 15 745 – 15 753

    2021

  49. [58]

    Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,

    K. Hundman, V . Constantinou, C. Laporte, I. Colwell, and T. Soder- strom, “Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,” in Proceedings of the 24th acm sigkdd inter- national conference on knowledge discovery & data mining , 2018, pp. 387–395

    2018

  50. [59]

    Practical approach to asyn- chronous multivariate time series anomaly detection and localization,

    A. Abdulaal, Z. Liu, and T. Lancewicki, “Practical approach to asyn- chronous multivariate time series anomaly detection and localization,” in Proceedings of the 27th acm sigkdd conference on knowledge discovery & data mining , 2021, pp. 2485–2494

    2021

  51. [60]

    Swat: A water treatment testbed for research and training on ics security,

    A. P. Mathur and N. O. Tippenhauer, “Swat: A water treatment testbed for research and training on ics security,” in2016 international workshop on cyber-physical systems for smart water networks , 2016, pp. 31–36

    2016

  52. [61]

    Lof: Identifying density- based local outliers,

    M. Breunig, H. Kriegel, R. Ng, and J. Sander, “Lof: Identifying density- based local outliers,” International conference on management of data , vol. 29, no. 2, pp. 93–104, 2000

    2000

  53. [62]

    Deep autoencoding gaussian mixture model for unsupervised anomaly detection,

    B. Zong, Q. Song, M. R. Min, W. Cheng, C. Lumezanu, D. Cho, and H. Chen, “Deep autoencoding gaussian mixture model for unsupervised anomaly detection,” in Proceedings of 6th international conference on learning representations, 2018

    2018

  54. [63]

    Auto-encoding variational bayes,

    D. P. Kingma and M. Welling, “Auto-encoding variational bayes,” in Proceedings of 2nd international conference on learning representations, 2014

    2014

  55. [64]

    Robust anomaly detection for multivariate time series through stochastic recurrent neural network,

    Y . Su, Y . Zhao, C. Niu, R. Liu, W. Sun, and D. Pei, “Robust anomaly detection for multivariate time series through stochastic recurrent neural network,” in Proceedings of the 25th acm sigkdd international confer- encce on knowledge discovery and data mining , 2019, pp. 2828–2837

    2019

  56. [65]

    Tranad: Deep transformer networks for anomaly detection in multivariate time series data,

    S. Tuli, G. Casale, and N. R. Jennings, “Tranad: Deep transformer networks for anomaly detection in multivariate time series data,” in Proceedings of VLDB endowment , vol. 15, no. 6, 2022, p. 1201–1214

    2022

  57. [66]

    Dcff-mtad: A multivariate time- series anomaly detection model based on dual-channel feature fusion,

    Z. Xu, Y . Yang, X. Gao, and M. Hu, “Dcff-mtad: A multivariate time- series anomaly detection model based on dual-channel feature fusion,” Sensors, vol. 23, no. 8, 2023

    2023

  58. [67]

    Memory-augmented u-transformer for multivariate time series anomaly detection,

    S. Qin, Y . Luo, and G. Tao, “Memory-augmented u-transformer for multivariate time series anomaly detection,” in Proceedings of IEEE international conference on acoustics, speech and signal processing , 2023

    2023

  59. [68]

    Btad: A binary transformer deep neural network model for anomaly detection in multivariate time series data,

    M. Ma, L. Han, and C. Zhou, “Btad: A binary transformer deep neural network model for anomaly detection in multivariate time series data,” Advanced Engineering Informatics , vol. 56, p. 101949, 2023

    2023

  60. [69]

    Deep temporal graph clustering,

    M. Liu, Y . Liu, K. Liang, W. Tu, S. Wang, S. Zhou, and X. Liu, “Deep temporal graph clustering,” in The 12th International Conference on Learning Representations, 2024

    2024

  61. [70]

    Self-supervised temporal graph learning with temporal and structural intensity alignment,

    M. Liu, K. Liang, Y . Zhao, W. Tu, S. Zhou, X. Gan, X. Liu, and H. Kunlun, “Self-supervised temporal graph learning with temporal and structural intensity alignment,” IEEE Transactions on Neural Networks and Learning Systems , 2024

    2024

  62. [71]

    Tmac: Temporal multi-modal graph learning for acoustic event classification,

    M. Liu, K. Liang, D. Hu, H. Yu, Y . Liu, L. Meng, W. Tu, S. Zhou, and X. Liu, “Tmac: Temporal multi-modal graph learning for acoustic event classification,” in Proceedings of the 31st ACM International Conference on Multimedia, 2023, pp. 3365–3374

    2023

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.