Pith. sign in

REVIEW 3 major objections 6 minor 55 references

FedKAD learns shared low-rank Koopman dynamics so edge IoT devices can detect multivariate time-series anomalies without shipping raw data or training large neural models.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-13 01:16 UTC pith:T36GVFY7

load-bearing objection Solid edge-FL systems paper: the efficiency story is real and well measured; the accuracy claim is competitive, not dominant, and rests on a fixed random lifting that is under-justified. the 3 major comments →

arxiv 2607.08978 v1 pith:T36GVFY7 submitted 2026-07-09 cs.LG eess.SP

Federated Low-Rank Koopman Learning for Multivariate Time-Series Anomaly Detection in IoT Systems

Tung-Anh Nguyen , Van-Phuc Bui , Anh Tuyen Le , Kim Hue Ta , Minh Thuy Le , J.Andrew Zhang , Xiaojing Huang This is my paper
classification cs.LG eess.SP
keywords Internet of ThingsFederated LearningAnomaly DetectionMultivariate Time SeriesKoopman OperatorStiefel ManifoldEdge IntelligenceADMM
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Distributed IoT sensors produce multivariate time series that must be monitored for faults and attacks, but the data sit on weak devices with non-IID streams and tight bandwidth. This paper argues that normal temporal behavior can be captured by a lightweight sliding-window Koopman model: a shared low-rank orthonormal subspace plus private reduced operators, rather than by federated deep networks. Only the compact subspace matrices are exchanged; raw streams and local dynamics stay on device. A federated Stiefel-ADMM procedure aligns those subspaces under partial client participation, with convergence and stationarity guarantees. On four standard benchmarks the method matches or beats federated neural detectors on segment-level F1 while cutting training time, communication volume, and inference latency by one to three orders of magnitude, including on a Raspberry Pi 4.

Core claim

The authors claim that federated multivariate time-series anomaly detection for constrained IoT can be solved by low-rank Koopman consensus: clients keep snapshot matrices and reduced operators private, agree only on a shared orthonormal subspace of the lifted windowed states, and flag anomalies by multi-step prediction residual. This formulation yields detection quality competitive with federated deep baselines while making training, communication, and on-device inference practical for edge hardware.

What carries the argument

Federated Stiefel-ADMM on the OMD-style factorization Φ(z(t+1)) ≈ P Q Pᵀ Φ(z(t)), with P on the Stiefel manifold and Q client-private: clients exchange only compact subspace matrices P, the server forms a consensus Z, and dual variables enforce agreement under partial participation.

Load-bearing premise

A fixed random tanh lifting plus one shared low-rank orthonormal subspace (with private reduced operators) is expressive enough that prediction residual reliably separates normal from abnormal dynamics across non-IID clients.

What would settle it

On the same four federated partitions, replace the fixed random lifting and rank-r shared subspace with a client-specific full-rank operator or a learned neural lifting of equal communication budget; if residual-based F1 under PA%K collapses relative to FedKAD while communication stays comparable, the expressiveness claim fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper proposes FedKAD, a federated anomaly detector for multivariate IoT time series that replaces heavy neural models with sliding-window Koopman dynamics. Each client lifts windowed states with a fixed shared random tanh map, fits a private reduced operator Qi, and participates in consensus on a shared orthonormal subspace Z via a federated Stiefel-ADMM algorithm; anomalies are scored by multi-step prediction residual. The authors give convergence and stationarity analysis under partial participation (Assumptions A1–A5, Theorems V.5 and V.7), evaluate on SMD, PSM, SMAP, and MSL under non-IID partitions and multiple thresholding protocols, and report competitive primary PA%K F1 (best on 3/4 datasets) together with large reductions in training time, communication, and inference latency, including a Raspberry Pi 4 deployment study.

Significance. If the empirical trade-off holds, this is a practically useful contribution for edge IoT anomaly detection: it shows that a low-rank dynamical model with compact subspace consensus can match or beat federated deep baselines on standard MVTS benchmarks while being orders of magnitude cheaper to train and communicate. Strengths include an explicit non-IID characterization (Fig. 2), multi-protocol evaluation motivated by known PA pathologies, a Stiefel-ADMM formulation with partial-participation analysis, and concrete on-device measurements rather than only GPU wall-clock. The work is more systems- and deployment-oriented than a pure theory paper, but the efficiency evidence and honest multi-protocol tables make it a credible candidate for an IoT/edge-learning venue.

major comments (3)
  1. Sec. IV-A–C and Eqs. (7)–(11): the accuracy half of the central claim rests on a fixed random tanh lifting plus a shared rank-r subspace with private Qi being expressive enough for residual scoring under the documented non-IID skew (Fig. 2). The manuscript reports no ablation of dlift, r (beyond a PSM validation choice), shared Z vs fully local subspaces, or residual histograms on normal validation data. Without these, it is hard to know when the representation fails (e.g., MSL under the primary protocol, and several losses in Table IV under stricter PA%K/point-wise settings). Please add targeted ablations and a short failure-mode discussion so the joint accuracy–efficiency claim is not tied only to the default hyperparameter point.
  2. Fig. 3 / Table III vs Table IV: the headline “best F1 on three of four datasets” is specific to PA%K with k=0.01. Under PA%K k=0.10 and strict point-wise no-PA, FedKAD loses on most datasets. The paper already reports these numbers, which is good, but the abstract and conclusion still emphasize “maintains or improves” without qualifying the protocol dependence. Please align abstract/conclusion wording with the multi-protocol evidence and state clearly that the main advantage is segment-level detection under mild PA%K rather than point-level localization.
  3. Sec. VI-B4 / Fig. 6: on-device inference latency is stated as “0.79 µs/step” in the text but “786 µs” in the figure annotation for FedKAD under the SMAP setting—a 1000× discrepancy. Relative speedups versus baselines may still hold, but absolute numbers and the claimed “79× lower inference latency” need to be reconciled with a single consistent unit system before the efficiency claim can be cited.
minor comments (6)
  1. Algorithm 1 is written with full client participation (“for client i = 1, …, N”), while Sec. IV-D and the theory use partial sets Sℓ and Eq. (16). Please make the algorithm match the partial-participation case used in experiments (25% sampling).
  2. Eq. (10) and the surrounding text: briefly state the condition under which Gi(Pi) is regularized (λ) and whether λ was tuned or fixed across datasets; a one-line sensitivity note would help reproducibility.
  3. Notation: both FEDKAD and FedKAD appear; pick one spelling. Also “FEDKAD” in Fig. 1 caption vs “FedKAD” elsewhere.
  4. Related work: a short comparison to other federated MVTS anomaly detectors (e.g., federated VAE / deep federated AD cited as [45]–[47]) on communication volume would situate the 80× claim more clearly.
  5. Fig. 5: late-round Lagrangian oscillation is attributed to εℓ, ΔQℓ, and partial participation; adding the participation schedule or |Sℓ| on the plot would make that interpretation easier to verify.
  6. Typos / polish: “It is easily to see that FEDKAD” (Sec. VI-B1); “PAon three of the four datasets” appears truncated in the main-results paragraph; arXiv-style “D”Iot” / encoding artifacts in references should be cleaned for the camera-ready.

Circularity Check

0 steps flagged

No significant circularity: residual-based AD, Stiefel-ADMM theory, and empirical FL benchmarks are standard constructions, not tautologies or self-citation chains.

full rationale

FedKAD’s derivation chain is self-contained and non-circular. (1) The anomaly score is the multi-step prediction residual of a Koopman model fitted only on normal training windows (Eqs. 8–11, Sec. IV-F); that is the intended unsupervised residual construction, not a redefinition of labels or a fit renamed as a prediction of the same quantity. (2) The low-rank factorization PQP⊤ and fixed tanh random lifting are taken from external OMD/EDMD/universal-approximation literature (Wynn et al., Williams et al., Cybenko), not from an author uniqueness theorem or a prior self-ansatz. (3) Federated Stiefel-ADMM and the stationarity/convergence results (Lemmas V.2–V.3, Theorems V.5–V.7) are standard ADMM-plus-manifold descent arguments under explicit assumptions A1–A5; they do not smuggle the target claim into the premises. (4) Detection and efficiency claims are empirical comparisons against FedAvg-wrapped neural baselines on four public MVTS benchmarks with multi-protocol metrics; hyperparameter choices (e.g., r on PSM) use held-out validation, which is ordinary model selection. No load-bearing step reduces by construction to its own inputs, and there is no self-citation uniqueness chain. Score 0 is appropriate.

Axiom & Free-Parameter Ledger

6 free parameters · 4 axioms · 1 invented entities

The central efficiency-plus-accuracy claim rests on modeling choices (fixed random lifting, shared rank-r subspace, private Q_i), standard optimization assumptions for inexact ADMM on Stiefel, and several hand-chosen hyperparameters. No new physical entities are postulated; the method is an algorithmic construct evaluated on public data.

free parameters (6)
  • subspace rank r = 24 (SMD/SMAP/MSL); 32 (PSM)
    Controls model capacity and communication; default 24, raised to 32 on PSM via held-out validation sweep.
  • lifting dimension d_lift = 128
    Size of fixed random nonlinear observables; default 128.
  • ADMM penalty ρ = 1.0
    Must be large enough for local strong convexity and consensus; default 1.0, ablated in Fig. 5.
  • Stiefel step size η and inner steps C = η=0.05
    Local manifold optimization hyperparameters; step size given as 0.05.
  • ridge λ and spectral cap = λ=1e-2; cap=0.995
    Stabilize Q_i fit and dynamics; λ=1e-2, spectral cap 0.995.
  • window W, horizon H, score smoothing = W=20, H=1
    Temporal context and residual aggregation; W=20, H=1, smoothing 5 steps.
axioms (4)
  • domain assumption Finite-dimensional Koopman approximation with fixed random tanh lifting is sufficiently expressive for normal MVTS dynamics on the studied benchmarks.
    Sec. IV-A, Eq. (7); lifting is not learned and is justified by universal approximation heuristics rather than a guarantee for these systems.
  • domain assumption Normal behavior admits a shared low-rank orthonormal subspace with client-specific reduced linear operators (OMD factorization).
    Eq. (8)–(11); core modeling choice enabling private Q_i and compact communication.
  • standard math A1–A5: Lipschitz Riemannian gradients, strong convexity of local Pi subproblems for large ρ, bounded Lagrangian, summable inexactness and Q-refit perturbations, and eventual full client participation over windows of rounds.
    Sec. V assumptions used for Theorems V.5 and V.7; standard for nonconvex ADMM analyses but not verified a priori on the data.
  • domain assumption Prediction residual of learned normal dynamics is a valid unsupervised anomaly score under the chosen thresholding protocols.
    Sec. IV-F and evaluation design; common in residual-based detectors but metric-dependent (PA%K vs point-wise).
invented entities (1)
  • FedKAD shared low-rank Koopman subspace consensus no independent evidence
    purpose: Coordinate a global orthonormal basis Z while keeping raw snapshots and reduced operators Q_i on device.
    Modeling construct introduced for federated efficiency; evaluated empirically, not an independent physical object.

pith-pipeline@v1.1.0-grok45 · 28044 in / 3244 out tokens · 38972 ms · 2026-07-13T01:16:14.473380+00:00 · methodology

0 comments
read the original abstract

Distributed IoT systems generate multivariate time-series streams for monitoring physical assets, servers, and embedded sensing platforms. Detecting abnormal temporal behavior is critical for fault diagnosis, predictive maintenance, and security. However, practical IoT anomaly detection is hindered by decentralized and non-IID data, limited bandwidth, and the constrained computation and memory of edge devices. This paper proposes FedKAD, a resource-efficient federated Koopman anomaly detection framework for distributed IoT multivariate time series. Unlike deep-learning-based anomaly detectors that require training and communicating large neural models, FedKAD learns normal temporal dynamics through lightweight sliding-window Koopman representations. Federated training is formulated as a low-rank consensus problem, where raw sensor streams and local reduced dynamics remain on device while only compact subspace variables are exchanged with the server. To optimize the shared representation under orthonormality constraints, we develop a federated Stiefel-ADMM algorithm and provide convergence and stationarity analysis under partial client participation. During inference, each client detects anomalies locally by measuring the prediction residual between observed future trajectories and the learned Koopman dynamics. Experiments on four widely used multivariate time-series anomaly detection benchmarks show that FedKAD maintains or improves detection performance compared with federated deep-learning baselines. More importantly for IoT deployment, FedKAD provides up to $2.1\times10^3$ faster training, $80\times$ lower communication, and $79\times$ lower inference latency than neural baselines, confirming its suitability for resource-constrained edge devices.

Figures

Figures reproduced from arXiv: 2607.08978 by Anh Tuyen Le, J.Andrew Zhang, Kim Hue Ta, Minh Thuy Le, Tung-Anh Nguyen, Van-Phuc Bui, Xiaojing Huang.

Figure 1
Figure 1. Figure 1: System model of FEDKAD for federated IoT multivariate time-series anomaly detection. subspace. The projection P ⊤ maps each raw snapshot to reduced coordinates, Q advances the reduced state, and P maps the prediction back to the original raw data space. Hence, P QP ⊤ ∈ R d×d provides a rank-r approximation of the finite￾dimensional transition operator in Eq. (2). E. ADMM-based Koopman Learning. The Alterna… view at source ↗
Figure 2
Figure 2. Figure 2: Non-IID heterogeneity across the four benchmarks, covering feature distribution skew, label distribution skew, and quantity skew. [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Performance under the primary PA%K protocol with [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Federated efficiency: training wall-clock (s), per-round communication (KB) and inference latency ( [PITH_FULL_IMAGE:figures/full_fig_p012_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Empirical convergence diagnostics on SMD ( [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: On-device benchmark on a Raspberry Pi 4 under the [PITH_FULL_IMAGE:figures/full_fig_p012_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

55 extracted references · 4 linked inside Pith

  1. [1]

    Anomaly detection: A survey,

    V . Chandola, A. Banerjee, and V . Kumar, “Anomaly detection: A survey,” ACM Computing Surveys, vol. 41, no. 3, pp. 15:1–15:58, 2009

    2009

  2. [2]

    Anomaly detection for iot time- series data: A survey,

    A. A. Cook, G. Misirli, and Z. Fan, “Anomaly detection for iot time- series data: A survey,”IEEE Internet of Things Journal, vol. 7, no. 7, pp. 6481–6494, 2020

    2020

  3. [3]

    Anomaly detection in streams with extreme value theory,

    A. Siffer, P.-A. Fouque, A. Termier, and C. Largou”et, “Anomaly detection in streams with extreme value theory,” inProceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2017, pp. 1067–1075

    2017

  4. [4]

    Unsupervised anomaly detection via variational auto- encoder for seasonal kpis in web applications,

    H. Xu, W. Chen, N. Zhao, Z. Li, J. Bu, Z. Li, Y . Liu, Y . Zhao, D. Pei, Y . Fenget al., “Unsupervised anomaly detection via variational auto- encoder for seasonal kpis in web applications,” inProceedings of the 2018 world wide web conference, 2018, pp. 187–196

    2018

  5. [5]

    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,” inICLR, 2018

    2018

  6. [6]

    Lstm-based encoder-decoder for multi-sensor anomaly detection,

    P. Malhotra, A. Ramakrishnan, G. Anand, L. Vig, P. Agarwal, and G. Shroff, “Lstm-based encoder-decoder for multi-sensor anomaly detection,” inICML 2016 Anomaly Detection Workshop, 2016, arXiv:1607.00148

    Pith/arXiv arXiv 2016

  7. [7]

    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,” inProceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining, 2019, pp. 2828– 2837

    2019

  8. [8]

    Multivariate time-series anomaly detection via graph attention network,

    H. Zhao, Y . Wang, J. Duan, C. Huang, D. Cao, Y . Tong, B. Xu, J. Bai, J. Tong, and Q. Zhang, “Multivariate time-series anomaly detection via graph attention network,” inICDM, 2020

    2020

  9. [9]

    Graph neural network-based anomaly detection in multivariate time series,

    A. Deng and B. Hooi, “Graph neural network-based anomaly detection in multivariate time series,” inAAAI, 2021

    2021

  10. [10]

    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,” inICLR, 2022

    2022

  11. [11]

    Tranad: Deep transformer net- works for anomaly detection in multivariate time series data,

    S. Tuli, G. Casale, and N. R. Jennings, “Tranad: Deep transformer net- works for anomaly detection in multivariate time series data,”Proceedings of the VLDB Endowment, vol. 15, no. 6, pp. 1201–1214, 2022

    2022

  12. [12]

    Unsupervised anomaly detection via variational auto-encoder for seasonal kpis in web applications,

    H. Xu, W. Chen, N. Zhao, Z. Li, J. Bu, Z. Li, Y . Liu, Y . Zhao, D. Pei, Y . Feng, J. Chen, Z. Wang, and H. Qiao, “Unsupervised anomaly detection via variational auto-encoder for seasonal kpis in web applications,” in Proceedings of the 2018 World Wide Web Conference, 2018, pp. 187–196

    2018

  13. [13]

    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,” inInternational Conference on Learning Representations, 2018. [Online]. Available: https: //openreview.net/forum?id=BJJLHbb0-

    2018

  14. [14]

    Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,

    K. Hundman, V . Constantinou, C. Laporte, I. Colwell, and T. Soderstrom, “Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,” inProceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2018, pp. 387– 395

    2018

  15. [15]

    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,” inProceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2019, pp. 2828– 2837

    2019

  16. [16]

    Multivariate time-series anomaly detection via graph attention network,

    H. Zhao, Y . Wang, J. Duan, C. Huang, D. Cao, Y . Tong, B. Xu, J. Bai, J. Tong, and Q. Zhang, “Multivariate time-series anomaly detection via graph attention network,” in2020 IEEE International Conference on Data Mining, 2020, pp. 841–850

    2020

  17. [17]

    Graph neural network-based anomaly detection in multivariate time series,

    A. Deng and B. Hooi, “Graph neural network-based anomaly detection in multivariate time series,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 35, no. 5, 2021, pp. 4027–4035

    2021

  18. [18]

    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,” inInternational Conference on Learning Representations, 2022. [Online]. Available: https://openreview.net/forum?id=LzQQ89U1qm

    2022

  19. [19]

    Communication-efficient learning of deep networks from de- centralized data,

    H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. Ag”uera y Arcas, “Communication-efficient learning of deep networks from de- centralized data,” inProceedings of the 20th International Conference on Artificial Intelligence and Statistics, ser. Proceedings of Machine Learning Research, vol. 54, 2017, pp. 1273–1282

    2017

  20. [20]

    Advances and open problems in federated learning,

    P. Kairouz, H. B. McMahan, B. Avent, A. Bellet, M. Bennis, A. N. Bhagoji, K. Bonawitz, Z. Charles, G. Cormode, R. Cummingset al., “Advances and open problems in federated learning,”Foundations and Trends in Machine Learning, vol. 14, no. 1–2, pp. 1–210, 2021

    2021

  21. [21]

    Federated learning for internet of things: A comprehensive survey,

    D. C. Nguyen, M. Ding, P. N. Pathirana, A. Seneviratne, J. Li, and H. V . Poor, “Federated learning for internet of things: A comprehensive survey,”IEEE Communications Surveys & Tutorials, vol. 23, no. 3, pp. 1622–1658, 2021

    2021

  22. [22]

    D”Iot: A federated self-learning anomaly detection system for iot,

    T. D. Nguyen, S. Marchal, M. Miettinen, H. Fereidooni, N. Asokan, and A.-R. Sadeghi, “D”Iot: A federated self-learning anomaly detection system for iot,” in2019 IEEE 39th International Conference on Distributed Computing Systems, 2019, pp. 756–767

    2019

  23. [23]

    Cross-node federated graph neural network for spatio-temporal data modeling,

    C. Meng, S. Rambhatla, and Y . Liu, “Cross-node federated graph neural network for spatio-temporal data modeling,” inProceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 2021, pp. 1202–1211

    2021

  24. [24]

    Hamiltonian systems and transformation in hilbert space,

    B. O. Koopman, “Hamiltonian systems and transformation in hilbert space,”Proceedings of the National Academy of Sciences of the United States of America, vol. 17, no. 5, pp. 315–318, 1931

    1931

  25. [25]

    Spectral analysis of nonlinear flows,

    C. W. Rowley, I. Mezi’c, S. Bagheri, P. Schlatter, and D. S. Henningson, “Spectral analysis of nonlinear flows,”Journal of Fluid Mechanics, vol. 641, pp. 115–127, 2009

    2009

  26. [26]

    A data-driven approximation of the koopman operator: Extending dynamic mode decomposition,

    M. O. Williams, I. G. Kevrekidis, and C. W. Rowley, “A data-driven approximation of the koopman operator: Extending dynamic mode decomposition,”Journal of Nonlinear Science, vol. 25, no. 6, pp. 1307– 1346, 2015. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 14

    2015

  27. [27]

    Dynamic mode decomposition of numerical and experi- mental data,

    P. J. Schmid, “Dynamic mode decomposition of numerical and experi- mental data,”Journal of Fluid Mechanics, vol. 656, pp. 5–28, 2010

    2010

  28. [28]

    Optimal mode decomposition for unsteady flows,

    A. Wynn, D. S. Pearson, B. Ganapathisubramani, and P. J. Goulart, “Optimal mode decomposition for unsteady flows,”Journal of Fluid Mechanics, vol. 733, pp. 473–503, 2013

    2013

  29. [29]

    Absil, R

    P.-A. Absil, R. Mahony, and R. Sepulchre,Optimization Algorithms on Matrix Manifolds. Princeton, NJ: Princeton University Press, 2008

    2008

  30. [30]

    Distributed optimization and statistical learning via the alternating direction method of multipliers,

    S. Boyd, N. Parikh, E. Chu, B. Peleato, J. Ecksteinet al., “Distributed optimization and statistical learning via the alternating direction method of multipliers,”Foundations and Trends® in Machine learning, vol. 3, no. 1, pp. 1–122, 2011

    2011

  31. [31]

    Towards a rigorous evaluation of time-series anomaly detection,

    S. Kim, K. Choi, H.-S. Choi, B. Lee, and S. Yoon, “Towards a rigorous evaluation of time-series anomaly detection,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 7, 2022, pp. 7194–7201

    2022

  32. [32]

    Federated learning: Challenges, methods, and future directions,

    T. Li, A. K. Sahu, A. Talwalkar, and V . Smith, “Federated learning: Challenges, methods, and future directions,”IEEE Signal Processing Magazine, vol. 37, no. 3, pp. 50–60, 2020

    2020

  33. [33]

    Fedpaq: A communication-efficient federated learning method with periodic averaging and quantization,

    A. Reisizadeh, A. Mokhtari, H. Hassani, A. Jadbabaie, and R. Pedarsani, “Fedpaq: A communication-efficient federated learning method with periodic averaging and quantization,” inProceedings of the 23rd International Conference on Artificial Intelligence and Statistics, 2020, pp. 2021–2031

    2020

  34. [34]

    Robust and communication-efficient federated learning from non-iid data,

    F. Sattler, S. Wiedemann, K.-R. M ¨uller, and W. Samek, “Robust and communication-efficient federated learning from non-iid data,” inIEEE Transactions on Neural Networks and Learning Systems, 2019

    2019

  35. [35]

    Expanding the reach of federated learning by reducing client resource requirements,

    S. Caldas, J. Kone ˇcn´y, H. B. McMahan, and A. Talwalkar, “Expanding the reach of federated learning by reducing client resource requirements,” inNeurIPS Workshop on Federated Learning for Data Privacy and Confidentiality, 2018

    2018

  36. [36]

    Long short term memory networks for anomaly detection in time series,

    P. Malhotra, L. Vig, G. Shroff, and P. Agarwal, “Long short term memory networks for anomaly detection in time series,” inESANN, 2015

    2015

  37. [37]

    Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,

    K. Hundman, V . Constantinou, C. Laporte, I. Colwell, and T. Soderstrom, “Detecting spacecraft anomalies using lstms and nonparametric dynamic thresholding,” inProceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, 2018, pp. 387–395

    2018

  38. [38]

    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,” inKDD, 2019

    2019

  39. [39]

    Mad-gan: Multivariate anomaly detection for time series data with generative adversarial networks,

    D. Li, D. Chen, B. Jin, L. Shi, J. Goh, and S.-K. Ng, “Mad-gan: Multivariate anomaly detection for time series data with generative adversarial networks,” inICANN, 2019

    2019

  40. [40]

    Tadgan: Time series anomaly detection using generative adversarial networks,

    A. Geiger, D. Liu, S. Alnegheimish, A. Cuesta-Infante, and K. Veera- machaneni, “Tadgan: Time series anomaly detection using generative adversarial networks,” inBigData, 2020

    2020

  41. [41]

    Informer: Beyond efficient transformer for long sequence time-series forecasting,

    H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, “Informer: Beyond efficient transformer for long sequence time-series forecasting,” inAAAI, 2021

    2021

  42. [42]

    Deep learning for universal linear embeddings of nonlinear dynamics,

    B. Lusch, J. N. Kutz, and S. L. Brunton, “Deep learning for universal linear embeddings of nonlinear dynamics,”Nature communications, vol. 9, no. 1, p. 4950, 2018

    2018

  43. [43]

    Deep koopman predictors for anomaly detection of complex iot systems with time series data,

    L. Fu, M. Ma, and Z. Zhai, “Deep koopman predictors for anomaly detection of complex iot systems with time series data,”IEEE Internet of Things Journal, vol. 11, no. 23, pp. 38 360–38 369, 2024

    2024

  44. [44]

    Model- free geometric fault detection and isolation for nonlinear systems using koopman operator,

    M. Bakhtiaridoust, M. Yadegar, N. Meskin, and M. Noorizadeh, “Model- free geometric fault detection and isolation for nonlinear systems using koopman operator,”IEEE Access, vol. 10, pp. 14 835–14 845, 2022

    2022

  45. [45]

    Federated variational learning for anomaly detection in multivariate time series,

    K. Zhang, Y . Jiang, L. Seversky, C. Xu, D. Liu, and H. Song, “Federated variational learning for anomaly detection in multivariate time series,” arXiv preprint arXiv:2108.08404, 2021

    Pith/arXiv arXiv 2021

  46. [46]

    Deep federated anomaly detection for multivariate time series data,

    W. Zhu, D. Song, Y . Chen, W. Cheng, B. Zong, T. Mizoguchi, C. Lumezanu, H. Chen, and J. Luo, “Deep federated anomaly detection for multivariate time series data,”arXiv preprint arXiv:2205.04041, 2022

    Pith/arXiv arXiv 2022

  47. [47]

    Federated time-series anomaly detection benchmark,

    F. Liuet al., “Federated time-series anomaly detection benchmark,”arXiv preprint arXiv:2212.09518, 2022

    Pith/arXiv arXiv 2022

  48. [48]

    Exploiting shared representations for personalized federated learning,

    L. Collins, H. Hassani, A. Mokhtari, and S. Shakkottai, “Exploiting shared representations for personalized federated learning,” inInternational conference on machine learning. PMLR, 2021, pp. 2089–2099

    2021

  49. [49]

    Data-driven discovery of koopman eigenfunctions for control,

    E. Kaiser, J. N. Kutz, and S. L. Brunton, “Data-driven discovery of koopman eigenfunctions for control,”Machine Learning: Science and Technology, vol. 2, no. 3, p. 035023, 2021

    2021

  50. [50]

    A data–driven approximation of the koopman operator: Extending dynamic mode decomposition,

    M. O. Williams, I. G. Kevrekidis, and C. W. Rowley, “A data–driven approximation of the koopman operator: Extending dynamic mode decomposition,”Journal of Nonlinear Science, vol. 25, no. 6, pp. 1307– 1346, 2015

    2015

  51. [51]

    Approximation by superpositions of a sigmoidal function,

    G. Cybenko, “Approximation by superpositions of a sigmoidal function,” Mathematics of control, signals and systems, vol. 2, no. 4, pp. 303–314, 1989

    1989

  52. [52]

    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,” inProceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, 2021, pp. 2485–2494

    2021

  53. [53]

    Deep one-class classification,

    L. Ruff, R. Vandermeulen, N. Goernitz, L. Deecke, S. A. Siddiqui, A. Binder, E. M ¨uller, and M. Kloft, “Deep one-class classification,” in Proceedings of the 35th International Conference on Machine Learning, ser. Proceedings of Machine Learning Research, J. Dy and A. Krause, Eds., vol. 80. PMLR, 10–15 Jul 2018, pp. 4393–4402. [Online]. Available: https:...

    2018

  54. [54]

    Usad: Unsupervised anomaly detection on multivariate time series,

    J. Audibert, P. Michiardi, F. Guyard, S. Marti, and M. A. Zuluaga, “Usad: Unsupervised anomaly detection on multivariate time series,” in Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, 2020, pp. 3395–3404

    2020

  55. [55]

    Anomaly detection in streams with extreme value theory,

    A. Siffer, P.-A. Fouque, A. Termier, and C. Largouet, “Anomaly detection in streams with extreme value theory,” inProceedings of the 23rd ACM SIGKDD international conference on knowledge discovery and data mining, 2017, pp. 1067–1075

    2017