Pith. sign in

REVIEW 4 major objections 5 minor 46 references

Federated Motor Imagery Classification for Privacy-Preserving Brain-Computer Interfaces

T0 review · 4 major / 5 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 Federated EEG motor-imagery classification can beat centralized training while keeping raw brain data on users' devices.

desk verdict Solid empirical FL-for-EEG paper with a real test-time batch dependency and an overclaimed centralized comparison; worth reviewing, needs revision. read the letter →

arxiv 2412.01079 v1 pith:R6E5UOLU submitted 2024-12-02 cs.LG cs.HC

Tianwang Jia , Lubin Meng , Siyang Li , Jiajing Liu , Dongrui Wu This is my paper · ORCID
classification cs.LGcs.HC
keywords brain-computerinterfacemotorimageryclassificationfederatedlearningEEGprivacybatchnormalizationsharpness-awareminimizationcross-subjectgeneralization
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 proposes FedBS, a federated learning method for EEG motor imagery classification that trains a global classifier across many users without moving raw EEG data to a server. FedBS combines local batch-specific batch normalization with a sharpness-aware minimization optimizer, so each client's batch-normalization layer stays private while the server obtains a complete, deployable model. In experiments on three public datasets with three deep-learning backbones, FedBS outperformed six existing federated approaches and, in the large majority of dataset–model combinations, also scored above centralized training that pools all data with no privacy protection. If these results are correct, they indicate that privacy-preserving collaboration among BCI users need not sacrifice accuracy and may even improve cross-subject generalization.

What carries the argument

The central mechanism is the combination of two named ingredients. First, local batch-specific batch normalization: each client trains its own BN affine parameters (\gamma, \$\beta$), which are never sent to the server, and only the non-BN parameters are aggregated and redistributed; instead of global running statistics, FedBS recomputes the BN mean \mu_B and variance \$sigma_B^{2}$ from each batch at both training and test time (Eqs. (1)-(2)), aligning feature distributions across subjects and letting the server model adapt to a new test subject. Second, the SAM optimizer: during local updates, each client first takes a gradient-ascent step of size \rho to locate a nearby point of higher loss and then descends from there (Eq. (8)), steering the model toward a flat minimum; when such models are averaged, the global model generalizes better. Together these components reduce feature shift among clients and improve cross-subject decoding accuracy.

What would settle it

Evaluate FedBS in streaming mode with one trial at a time, so no batch exists for computing BN statistics; if accuracy collapses to chance level, the empirical advantage over centralized training is contingent on batch availability. A concrete check would be reproducing the three-dataset experiments with a test batch size of 1, or showing that a causal running average of past trials preserves the reported performance.

Watch

Extended reading notes

Core claim

In the paper's own terms, the discovery is that a federated classifier can be simultaneously private and more accurate than pooled-data training by keeping batch-normalization parameters local and computing batch-specific statistics at test time, while each client trains with the SAM optimizer. FedBS gives the server a complete model: the server aggregates all parameters but distributes only the non-BN weights, so each client's BN layer remains private; at inference the BN mean and variance are recomputed from the current test batch, adapting the model to a new subject's distribution. The experiments report consistent gains over FedAvg, FedProx, SCAFFOLD, MOON, FedFA, and GA across EEGNet, DeepConvNet, and ShallowConvNet, with statistically significant improvements in most comparisons and with ablation studies confirming that both components contribute.

Load-bearing premise

The load-bearing premise is that the server holds a batch of test trials from the new user so it can compute batch-specific batch-normalization statistics at inference; the paper gives no single-trial classification procedure, so in a real-time BCI where trials arrive one at a time this assumption must be handled separately.

Editorial extensions

If this is right

  • If FedBS is correct, a hospital or BCI vendor can train a deployment-ready classifier across many users' data without collecting raw EEG, and that classifier should classify a new user at least as well as one trained on centrally pooled data.
  • The gains are not tied to a single network: FedBS worked with EEGNet, DeepConvNet, and ShallowConvNet, so it can be inserted into existing EEG deep-learning pipelines.
  • Unlike FedBN, FedBS leaves the server with a complete, immediately usable classifier rather than a personalized per-client model, which is what enables cross-subject testing.
  • Test-time batch size is a practical design parameter: accuracy rises with batch size and levels off around four trials per test batch, so the server needs only a small batch of trials from the new subject.

Reading between the lines

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

  • A natural extension would be to replace the required test batch with a short causal buffer of the new subject's own past trials, enabling single-trial real-time BCI operation; this would make FedBS directly usable in streaming settings.
  • The same recipe—client-private normalization with batch-specific statistics plus flat-minima optimization—could transfer to other federated medical or wearable domains where each device has its own distribution and the server must serve new users.
  • The result that privacy-preserving federated training can beat centralized pooling is surprising under the usual assumption that more data always helps; it suggests that pooled training may suffer from inter-subject distribution shift that batch-specific normalization naturally counteracts, which deserves a dedicated matched-hyperparameter study.
  • The ablation shows each component helps separately, so a finer parsing of the contribution of normalization versus flatness could let practitioners adopt only the component their infrastructure supports.
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

4 major / 5 minor

Summary. This paper proposes FedBS, a federated learning algorithm for EEG motor-imagery classification that combines local batch-specific batch normalization (BN) with sharpness-aware minimization (SAM). In FedBS, each client's BN parameters remain local and are not uploaded to the server, while the server aggregates the remaining parameters; at inference time, BN statistics are computed from a test batch. The authors evaluate FedBS against centralized training and six FL baselines (FedAvg, FedProx, SCAFFOLD, MOON, FedFA, GA) on three public MI datasets with three deep-learning backbones, report cross-subject accuracies, perform paired t-tests with Benjamini-Hochberg correction, and include ablation studies. The code is released. The central claim is that FedBS outperforms all compared FL approaches and usually also outperforms centralized training while preserving data privacy.

Significance. The core idea is a plausible and simple combination of two existing techniques (local BN and SAM) applied to FL for EEG, and the paper offers a broad experimental comparison on public benchmarks with three architectures, ablations, and publicly available code. If the claims are fully supported, FedBS would be a useful baseline for privacy-preserving FL in BCI research. However, the current evaluation is offline only and depends on a nonstandard test-time batch, the comparison with baselines is confounded by the test-time normalization protocol, and the statistical support contains a serious pseudoreplication issue, so the significance of the contribution as stated is not yet established.

major comments (4)
  1. [Section III-C, IV-C, IV-H] The method has no defined single-trial inference path. Equations (1) and (2) compute BN statistics from a test batch; Section IV-C fixes the test batch size to 8, and Fig. 6 shows accuracy depends on batch size and converges only at size 4. For an online BCI, trials arrive one at a time and a batch of size 1 has undefined variance. The paper gives no procedure for streaming inference (e.g., using running statistics from a calibration set), so the abstract's claim that FedBS improves BCI decoding accuracy is only supported for offline batch classification. Please add and evaluate a single-trial inference protocol, or revise the claims to clearly state the offline setup.
  2. [Section III-C and IV-C] The comparison between FedBS and the baselines is not controlled: FedBS uses batch-specific BN statistics at test time (Eqs. 1-2), whereas the baselines and CT are described as using standard batch normalization, which normally employs running statistics accumulated during training. The accuracy advantage of FedBS could therefore come from the test-time adaptation rather than from the federated training procedure. To support the central claim, the authors should evaluate all methods under the same inference-time normalization protocol, e.g., by also applying batch-specific statistics to the baselines, or by using running statistics for FedBS.
  3. [Appendix A, Tables X-XII] The reported degrees of freedom (53, 83, 71) imply that each of the six random-seed repeats per subject was treated as an independent observation. Since accuracies from the same test subject across repeats are correlated, this is pseudoreplication that inflates the sample size and makes the p-values anti-conservative. The significance tests should be performed on subject-level averaged accuracies (df = n_subjects - 1) or with a mixed-effects model that accounts for subject as a random factor, and the conclusions should be adjusted accordingly.
  4. [Abstract and Section IV-D, Table VII] The abstract's claim that FedBS outperformed centralized training is not supported for all settings: in Table VII, FedBS vs CT is not significant for ShallowConvNet on MI2 (p=0.313) and MI3 (p=0.206), and only borderline on MI1 (p=0.050); for DeepConvNet on MI2, p=0.219. The unqualified statement about outperforming six FL approaches also ignores several non-significant comparisons (e.g., ShallowConvNet on MI3 vs FedAvg, p=0.059, and vs FedProx, p=0.082). Please qualify the claims to reflect the statistical results.
minor comments (5)
  1. [Table VIII] The 'BN' and 'SAM' columns contain garbled symbols (e.g., '/enc-37 /enc-33'), which appear to be check/cross marks that failed to render; please replace them with proper glyphs and ensure the table is readable.
  2. [Tables III and Section IV-B] There are inconsistent spellings: 'DeepConvnet' in Table III should be 'DeepConvNet', and 'FedFa' in Section IV-B should be 'FedFA'.
  3. [Algorithm 1 and Section IV-C] Please define the client selection weight P explicitly and state its value in the main experiments; Section IV-C says half of the clients are selected (rounded down), but the mapping to P is not stated, and the expression max(P·K, 1) needs a rounding convention.
  4. [Fig. 6] Please add axis labels to Fig. 6 and describe how 'converged at 4' was determined (e.g., the threshold on accuracy difference from larger batch sizes).
  5. [Section IV-D] The numerical summary 'FedBS outperformed CT by 1.97%' should be accompanied by the qualification that several per-model comparisons were not statistically significant, as detailed in Table VII.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FedBS is benchmarked on held-out test subjects against independent federated baselines and centralized training; the disclosed test-batch-size dependence is a practical limitation, not a fitted prediction.

full rationale

FedBS's derivation chain is self-contained. The method is defined by local batch-specific batch normalization (Eqs. 1-2) plus the SAM objective (Eqs. 3-8), neither of which encodes the reported accuracies. Performance is measured with leave-one-subject-out cross-validation on three public MI datasets, compared with six external FL baselines and centralized training; no parameter of FedBS is fitted to the test outcomes. Author self-citations appear only in background/related work (e.g., refs. [3], [4], [17]-[20], [41], [46]) and are not load-bearing: the no-EA results in Table III already show FedBS best in every dataset-model column, so the same-author Euclidean-alignment preprocessing used in Tables IV-VI cannot explain the central claim. The paper does disclose a real test-time dependency, stating in Section IV-H that 'the batch size also impacts the test results' and that performance 'increased with the test batch size, but converged at 4'; this is a genuine limitation for single-trial online BCI inference, but it is not circularity because the reported comparison is an empirical result under a stated batch protocol rather than a quantity forced by definition or by a self-citation. No equation in the paper equates an output to an input, and no load-bearing premise is justified solely by the authors' prior work. Therefore the appropriate finding is no significant circularity.

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

The central claim depends on several manually chosen hyperparameters (rho, test batch size, P, E) and on two domain assumptions: that FL alone guarantees privacy, and that batched test-time normalization is a realistic inference setting. No new physical or formal entities are introduced.

free parameters (4)
  • SAM perturbation radius rho = 0.1
    Set in Section IV-C; no sensitivity analysis provided; it controls the sharpness-aware update and affects generalization.
  • Test batch size = 8
    All reported experiments use test batch size 8; performance varies with batch size (Fig. 6), and single-trial inference is not supported.
  • Client selection weight P = 0.5
    Half of clients selected per round in the main experiments; swept in Fig. 3 but fixed for the headline results.
  • Local epochs E = 2
    Two local epochs per communication round in the main experiments; swept in Fig. 4 but fixed for the headline results.
assumptions (3)
  • domain assumption Federated learning alone provides user privacy protection
    The paper relies on this in the Introduction and Section III; no privacy attack evaluation or defense such as differential privacy or secure aggregation is provided.
  • ad hoc to paper Batch-specific BN statistics computed on a test batch of 8 trials yield a valid classifier for BCI decoding
    This is the method's inference protocol described in Section III-C and IV-H; it is not standard single-trial BCI operation.
  • domain assumption Leave-one-subject-out cross-validation with six random seeds gives reliable performance estimates
    Used for all comparisons; standard practice, but it assumes seed variability and subject variability are representative.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Federated Motor Imagery Classification for Privacy-Preserving Brain-Computer Interfaces." pith.science (2026). https://pith.science/paper/R6E5UOLU

@misc{pith2026241201079,
  author       = {Pith},
  title        = {Pith review of: Federated Motor Imagery Classification for Privacy-Preserving Brain-Computer Interfaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R6E5UOLU}},
  note         = {Machine review of arXiv:2412.01079}
}
read the original abstract

Training an accurate classifier for EEG-based brain-computer interface (BCI) requires EEG data from a large number of users, whereas protecting their data privacy is a critical consideration. Federated learning (FL) is a promising solution to this challenge. This paper proposes Federated classification with local Batch-specific batch normalization and Sharpness-aware minimization (FedBS) for privacy protection in EEG-based motor imagery (MI) classification. FedBS utilizes local batch-specific batch normalization to reduce data discrepancies among different clients, and sharpness-aware minimization optimizer in local training to improve model generalization. Experiments on three public MI datasets using three popular deep learning models demonstrated that FedBS outperformed six state-of-the-art FL approaches. Remarkably, it also outperformed centralized training, which does not consider privacy protection at all. In summary, FedBS protects user EEG data privacy, enabling multiple BCI users to participate in large-scale machine learning model training, which in turn improves the BCI decoding accuracy.

Figures

Figures reproduced from arXiv: 2412.01079 by the authors.

Figure 1
Figure 1. FedBS for privacy-preserving BCIs. FedAvg [7] is one of the most popular FL approaches. To reduce the communication overhead, FedAvg performs multiple stochastic gradient descent updates on some chose n clients in each communication round and then aggregates the models on the server until convergence. FedBN [8] uses local batch normalization (BN) [9] and excludes the BN layer parameters of all client models in commu… view at source ↗
Figure 2
Figure 2. Overview of FedBS. Each client represents an individ [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Average classification accuracies of different FL ap [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Average classification accuracies of different FL ap [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: t-SNE visualization of feature extracted from test Subject 1 on the MI2 dataset. (a) CT; (b) FedAvg; and, (c) FedBS [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Classification accuracies of FedBS under different t [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 36 canonical work pages

  1. [1]

    Brain–computer interfaces for communication an d control,

    J. R. Wolpaw, N. Birbaumer, D. J. McFarland, G. Pfurtsche ller, and T. M. V aughan, “Brain–computer interfaces for communication an d control,” Clinical Neurophysiology, vol. 113, no. 6, pp. 767–791, 2002

    work page 2002

  2. [2]

    Motor imagery and direc t brain- computer communication,

    G. Pfurtscheller and C. Neuper, “Motor imagery and direc t brain- computer communication,” Proc. of the IEEE , vol. 89, no. 7, pp. 1123– 1134, 2001

    2001

  3. [3]

    Transfer learning for motor imagery based brain-computer interfaces: A tutorial,

    D. Wu, X. Jiang, and R. Peng, “Transfer learning for motor imagery based brain-computer interfaces: A tutorial,” Neural Networks, vol. 153, pp. 235–253, 2022

    2022

  4. [4]

    Privacy-Preserving Brain-Com puter Interfaces: A Systematic Review,

    K. Xia, W. Duch, Y . Sun, K. Xu, W. Fang, H. Luo, Y . Zhang, D. S ang, X. Xu, F.-Y . Wang, and D. Wu, “Privacy-Preserving Brain-Com puter Interfaces: A Systematic Review,” IEEE Trans. on Computational Social Systems, vol. 10, no. 5, pp. 2312–2324, 2023

    work page 2023

  5. [5]

    Federated machine l earning: Concept and applications,

    Q. Y ang, Y . Liu, T. Chen, and Y . Tong, “Federated machine l earning: Concept and applications,” ACM Transactions on Intelligent Systems and Technology, vol. 10, no. 2, pp. 1–19, 2019

    work page 2019

  6. [6]

    A Survey on Federated Learning Systems: Vision, Hype and Re ality for Data Privacy and Protection,

    Q. Li, Z. Wen, Z. Wu, S. Hu, N. Wang, Y . Li, X. Liu, and B. He, “A Survey on Federated Learning Systems: Vision, Hype and Re ality for Data Privacy and Protection,” IEEE Trans. on Knowledge and Data Engineering, vol. 35, no. 4, pp. 3347–3366, 2023

    work page 2023

  7. [7]

    Communication-Efficient Learning of Deep Networks from De central- ized Data,

    B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. Arca s, “Communication-Efficient Learning of Deep Networks from De central- ized Data,” in Proc. of Int’l Conf. on Artificial Intelligence and Statisti cs, vol. 54, Ft. Lauderdale, FL, Apr. 2017, pp. 1273–1282

    work page 2017

  8. [8]

    FedBN: Fede rated Learning on Non-IID Features via Local Batch Normalization ,

    X. Li, M. Jiang, X. Zhang, M. Kamp, and Q. Dou, “FedBN: Fede rated Learning on Non-IID Features via Local Batch Normalization ,” in Proc. of Int’l Conf. on Learning Representations , Vienna, Austria, May 2021

    work page 2021

Show all 46 references
  1. [9]

    Batch normalization: Accelera ting deep network training by reducing internal covariate shift,

    S. Ioffe and C. Szegedy, “Batch normalization: Accelera ting deep network training by reducing internal covariate shift,” in Proc. of Int’l Conf. on Machine Learning , Lille, France, Jul. 2015, pp. 448–456

    2015

  2. [10]

    Shar pness-aware Minimization for Efficiently Improving Generalization,

    P . Foret, A. Kleiner, H. Mobahi, and B. Neyshabur, “Shar pness-aware Minimization for Efficiently Improving Generalization,” i n Proc. of Int’l Conf. on Learning Representations , Vienna, Austria, May 2021

    2021

  3. [11]

    Privacy-preserving mach ine learning: Threats and solutions,

    M. Al-Rubaie and J. M. Chang, “Privacy-preserving mach ine learning: Threats and solutions,” IEEE Security & Privacy , vol. 17, no. 2, pp. 49–58, 2019

    2019

  4. [12]

    Privacy-preserving machine learning: Methods, challenges and directions,

    R. Xu, N. Baracaldo, and J. Joshi, “Privacy-preserving machine learning: Methods, challenges and directions,” arXiv preprint arXiv:2108.04417 , 2021

    2021 arXiv

  5. [13]

    Privacy preserving extreme learning machin e using additively homomorphic encryption,

    S. Kuri, T. Hayashi, T. Omori, S. Ozawa, Y . Aono, L. T. Pho ng, L. Wang, and S. Moriai, “Privacy preserving extreme learning machin e using additively homomorphic encryption,” in Proc. of IEEE Symposium Series on Computational Intelligence , Honolulu, HI, Nov. 2017, pp. 1–8

    2017

  6. [14]

    Common randomness in infor mation theory and cryptography. I. Secret sharing,

    R. Ahlswede and I. Csiszar, “Common randomness in infor mation theory and cryptography. I. Secret sharing,” IEEE Trans. on Information Theory, vol. 39, no. 4, pp. 1121–1132, 1993

    1993

  7. [15]

    Protecting Privacy of Us ers in Brain-Computer Interface Applications,

    A. Agarwal, R. Dowsley, N. D. McKinney, D. Wu, C.-T. Lin, M. De Cock, and A. C. A. Nascimento, “Protecting Privacy of Us ers in Brain-Computer Interface Applications,” IEEE Trans. on Neural Systems and Rehabilitation Engineering , vol. 27, no. 8, pp. 1546–1555, 2019

    2019

  8. [16]

    Source data- absent unsupervised domain adaptation through hypothesis transf er and labeling transfer,

    J. Liang, D. Hu, Y . Wang, R. He, and J. Feng, “Source data- absent unsupervised domain adaptation through hypothesis transf er and labeling transfer,” IEEE Trans. on Pattern Analysis and Machine Intelligence , vol. 44, no. 11, pp. 8602–8617, 2021

    2021

  9. [17]

    Transfer learning for EEG-ba sed brain- computer interfaces: A review of progress made since 2016,

    D. Wu, Y . Xu, and B.-L. Lu, “Transfer learning for EEG-ba sed brain- computer interfaces: A review of progress made since 2016,” IEEE Trans. on Cognitive and Developmental Systems , vol. 14, no. 1, pp. 4–19, 2022

    2016

  10. [18]

    Privacy-preserving domain adaptation for motor imagery-based brain-computer interf aces,

    K. Xia, L. Deng, W. Duch, and D. Wu, “Privacy-preserving domain adaptation for motor imagery-based brain-computer interf aces,” IEEE Trans. on Biomedical Engineering, vol. 69, no. 11, pp. 3365–3376, 2022

    2022

  11. [19]

    Lightweight source-free transfer f or privacy- preserving motor imagery classification,

    W. Zhang and D. Wu, “Lightweight source-free transfer f or privacy- preserving motor imagery classification,” IEEE Trans. on Cognitive and Developmental Systems , early access, 2022

    2022

  12. [20]

    Multi-source decentraliz ed transfer for privacy-preserving BCIs,

    W. Zhang, Z. Wang, and D. Wu, “Multi-source decentraliz ed transfer for privacy-preserving BCIs,” IEEE Trans. on Neural Systems and Rehabilitation Engineering , vol. 30, pp. 2710–2720, 2022

    2022

  13. [21]

    Federated Optimization in Heterogeneous Networks,

    T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, an d V . Smith, “Federated Optimization in Heterogeneous Networks,” in Proc. of Ma- chine Learning and Systems, vol. 2, Austin, TX, Mar. 2020, pp. 429–450

    2020

  14. [22]

    Scaffold: Stochastic controlled averaging for fe derated learn- ing,

    S. P . Karimireddy, S. Kale, M. Mohri, S. Reddi, S. Stich, and A. T. Suresh, “Scaffold: Stochastic controlled averaging for fe derated learn- ing,” in Proc. of Int’l Conf. on Machine Learning , online, Jul. 2020, pp. 5132–5143

    2020

  15. [23]

    Measuring the effects of non- identical data distribution for federated visual classific ation,

    T.-M. H. Hsu, H. Qi, and M. Brown, “Measuring the effects of non- identical data distribution for federated visual classific ation,” arXiv preprint arXiv:1909.06335 , 2019

    1909 arXiv

  16. [24]

    Adaptive Federated Optimizati on,

    S. J. Reddi, Z. Charles, M. Zaheer, Z. Garrett, K. Rush, J . Koneˇ cn´ y, S. Kumar, and H. B. McMahan, “Adaptive Federated Optimizati on,” in Proc. of Int’l Conf. on Learning Representations , Vienna, Austria, May 2021

    2021

  17. [25]

    Accele rated federated learning with decoupled adaptive optimization,

    J. Jin, J. Ren, Y . Zhou, L. Lyu, J. Liu, and D. Dou, “Accele rated federated learning with decoupled adaptive optimization,” in Proc. of Int’l Conf. on Machine Learning , Baltimore, MD, Jul. 2022, pp. 10 298–10 322

    2022

  18. [26]

    Federa ted Transfer Learning for EEG Signal Classification,

    C. Ju, D. Gao, R. Mane, B. Tan, Y . Liu, and C. Guan, “Federa ted Transfer Learning for EEG Signal Classification,” in Proc. of Int’l Conf. of the IEEE Engineering in Medicine & Biology Society , online, Jul. 2020, pp. 3040–3045

    2020

  19. [27]

    Aggreg ating Intrinsic Information to Enhance BCI Performance through F ederated Learning,

    R. Liu, Y . Chen, A. Li, Y . Ding, H. Y u, and C. Guan, “Aggreg ating Intrinsic Information to Enhance BCI Performance through F ederated Learning,” arXiv preprint arXiv:2308.11636 , 2023

    2023 arXiv

  20. [28]

    Improving gen eralization in federated learning by seeking flat minima,

    D. Caldarola, B. Caputo, and M. Ciccone, “Improving gen eralization in federated learning by seeking flat minima,” in Proc. of European Conf. on Computer Vision , Tel Aviv, Israel, Oct. 2022, pp. 654–672

    2022

  21. [29]

    Generali zed federated learning via sharpness aware minimization,

    Z. Qu, X. Li, R. Duan, Y . Liu, B. Tang, and Z. Lu, “Generali zed federated learning via sharpness aware minimization,” in Proc. of Int’l Conf. on Machine Learning , Baltimore, MD, Jul. 2022, pp. 18 250– 18 280

    2022

  22. [30]

    Exceeding chance level by c hance: The caveat of theoretical chance levels in brain signal classifi cation and statistical assessment of decoding accuracy,

    E. Combrisson and K. Jerbi, “Exceeding chance level by c hance: The caveat of theoretical chance levels in brain signal classifi cation and statistical assessment of decoding accuracy,” Journal of Neuroscience Methods, vol. 250, pp. 126–136, 2015

    2015

  23. [31]

    Review of the BCI Competition IV,

    M. Tangermann, K.-R. M¨ uller, A. Aertsen, N. Birbaumer , C. Braun, C. Brunner, R. Leeb, C. Mehring, K. Miller, G. Mueller-Putz, G. Nolte, G. Pfurtscheller, H. Preissl, G. Schalk, A. Schl¨ ogl, C. Vid aurre, S. Waldert, and B. Blankertz, “Review of the BCI Competition IV,” Fro...

    2012

  24. [32]

    Random forests in non-invasive sensorimotor rhythm brain-comput er interfaces: a practical and convenient non-linear classifier,

    D. Steyrl, R. Scherer, J. Faller, and G. R. M¨ uller-Putz , “Random forests in non-invasive sensorimotor rhythm brain-comput er interfaces: a practical and convenient non-linear classifier,” Biomedical Engineering / Biomedizinische Technik , vol. 61, no. 1, pp. 77–86, 2016

    2016

  25. [33]

    Autocalibration and Recurrent Adaptation: Towards a Plug and Play Online ERD-BCI,

    J. Faller, C. Vidaurre, T. Solis-Escalante, C. Neuper, and R. Scherer, “Autocalibration and Recurrent Adaptation: Towards a Plug and Play Online ERD-BCI,” IEEE Trans. on Neural Systems and Rehabilitation Engineering, vol. 20, no. 3, pp. 313–319, 2012

    2012

  26. [34]

    MOABB: trustworthy algor ithm bench- marking for BCIs,

    V . Jayaram and A. Barachant, “MOABB: trustworthy algor ithm bench- marking for BCIs,” Journal of Neural Engineering , vol. 15, no. 6, p. 066011, 2018

    2018

  27. [35]

    Model-Contrastive Federated Learning,

    Q. Li, B. He, and D. Song, “Model-Contrastive Federated Learning,” in Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recognition, online, Jun. 2021

    2021

  28. [36]

    FedFA: Federated Feature Aug mentation,

    T. Zhou and E. Konukoglu, “FedFA: Federated Feature Aug mentation,” in Proc. of Int’l Conf. on Learning Representations , Kigali, Rwanda, May 2023

    2023

  29. [37]

    Fe derated domain generalization with generalization adjustment,

    R. Zhang, Q. Xu, J. Y ao, Y . Zhang, Q. Tian, and Y . Wang, “Fe derated domain generalization with generalization adjustment,” i n Proc. of the IEEE/CVF Conf. on Computer Vision and Pattern Recogniti on, V ancouver, Canada, Jun. 2023, pp. 3954–3963

    2023

  30. [38]

    EEGNet: a compact convolutional neural net work for EEG-based brain–computer interfaces,

    V . J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P . Hung, and B. J. Lance, “EEGNet: a compact convolutional neural net work for EEG-based brain–computer interfaces,” Journal of Neural Engineering , vol. 15, no. 5, p. 056013, 2018

    2018

  31. [39]

    R. T. Schirrmeister, J. T. Springenberg, L. D. J. Fieder er, M. Glasstetter, K. Eggensperger, M. Tangermann, F. Hutter, W. Burgard, and T . Ball, “Deep learning with convolutional neural networks for EEG d ecoding 11 TABLE X PAIRED t-TEST DETAILS BETWEEN FED BS AND OTHER APPRO...

    2017

  32. [40]

    Evaluation of hyperparam- eter optimization in machine and deep learning methods for d ecoding imagined speech eeg,

    C. Cooney, A. Korik, R. Folli, and D. Coyle, “Evaluation of hyperparam- eter optimization in machine and deep learning methods for d ecoding imagined speech eeg,” Sensors, vol. 20, no. 16, p. 4629, 2020

    2020

  33. [41]

    Transfer Learning for Brain-Computer I nterfaces: A Euclidean Space Data Alignment Approach,

    H. He and D. Wu, “Transfer Learning for Brain-Computer I nterfaces: A Euclidean Space Data Alignment Approach,” IEEE Trans. on Biomedical Engineering, vol. 67, no. 2, pp. 399–410, 2020

    2020

  34. [42]

    Controlling the false di scovery rate: a practical and powerful approach to multiple testing,

    Y . Benjamini and Y . Hochberg, “Controlling the false di scovery rate: a practical and powerful approach to multiple testing,” Journal of the Royal statistical society: series B (Methodological) , vol. 57, no. 1, pp. 289–300, 1995

    1995

  35. [43]

    Visualizing data using t-SNE,

    L. V an der Maaten and G. Hinton, “Visualizing data using t-SNE,” Journal of machine learning research , vol. 9, no. 11, pp. 2579–2605, 2008

    2008

  36. [44]

    Quantify- ing the separability of data classes in neural networks,

    A. Schilling, A. Maier, R. Gerum, C. Metzner, and P . Krau ss, “Quantify- ing the separability of data classes in neural networks,” Neural Networks, vol. 139, pp. 278–293, 2021

    2021

  37. [45]

    On the surp rising behavior of distance metrics in high dimensional space,

    C. C. Aggarwal, A. Hinneburg, and D. A. Keim, “On the surp rising behavior of distance metrics in high dimensional space,” in Proc. of Int’l Conf. on Database Theory , London, UK, Jan. 2001, pp. 420–434

    2001

  38. [46]

    Affective brain-com puter interfaces (aBCIs): A tutorial,

    D. Wu, B.-L. Lu, B. Hu, and Z. Zeng, “Affective brain-com puter interfaces (aBCIs): A tutorial,” Proc. of the IEEE , vol. 11, no. 10, pp. 1314–1332, 2023. APPENDIX A. Paired t-tests between FedBS and other approaches We conducted paired t-tests between FedBS and other approach...

    2023

Pith tools

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