Pith. sign in

REVIEW 4 major objections 5 minor 31 references

ESTM: An Enhanced Dual-Branch Spectral-Temporal Mamba for Anomalous Sound Detection

T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Dual-branch Mamba tops machine-anomaly sound detection at 95.76 AUC

desk verdict ESTM is a well-built dual-branch Mamba with a clean internal ablation, but its SOTA margin rests on a protocol-mismatched comparison and needs verification. read the letter →

arxiv 2509.02471 v1 pith:W5OEQCES submitted 2025-09-02 cs.SD cs.LG

Chengyuan Ma , Peng Jia , Hongyue Guo , Wenming Yang This is my paper · ORCID
classification cs.SDcs.LG
keywords anomaloussounddetectionMambaselectivestate-spacemodelspectral-temporalfusionself-supervisedclassificationMelspectrogramenhancementDCASE2020Task2time-frequencydecoupling
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 sets out to show that anomalous machine sounds can be detected more reliably when the detector models time and frequency separately with selective state-space (Mamba) layers instead of relying on local receptive fields. Its ESTM system feeds a fused spectrogram—raw Log-Mel, a statistically gated Mel version, and the raw-waveform Tgram—into two patch-based Mamba branches, one scanning along frequency and one along time, then trains the whole network as an ID classifier with ArcFace loss. On the DCASE 2020 Task 2 benchmark, ESTM is reported to reach an average AUC of 95.76% and pAUC of 90.49%, exceeding the best prior published average by 0.84 and 1.79 points. The significance for a general reader: this is a concrete step toward recognizing never-before-heard machine failures from normal-only recordings, a setting where long-term trends and cross-band energy shifts matter.

What carries the argument

The STMamba backbone is the central mechanism: two selective-state-space (Mamba) branches, SMamba and TMamba, scan patched spectrograms along the frequency axis and the time axis respectively, each using learnable class tokens and positional embeddings, then align and add their outputs. The supporting mechanism is the TriStat-Gating (TSG) module, which multiplies the Log-Mel spectrogram frame-wise by a sigmoid-gated combination of median, RMS, and variance statistics to produce the enhanced ESgram. Together they are meant to capture long-range temporal dependencies and cross-band energy coupling that local-receptive-field models miss.

What would settle it

Take the six methods in Table I and retrain each on the same normal-sound training set used for ESTM (development plus additional), then evaluate on the same development test split. If ESTM no longer holds at least a 0.84-point average-AUC lead, the headline claim is refuted. A second check: train ESTM on the development set only; if its average AUC drops near or below the published baselines, most of the advantage comes from extra training data rather than the architecture.

Watch

Extended reading notes

Core claim

The paper argues that anomalous sound detection improves when a spectrogram is processed by two separate selective state-space branches, one scanning frequency patches and the other scanning time patches, and their outputs are fused. Before the scan, the Log-Mel spectrogram is enhanced by a parameter-free TriStat-Gating module that multiplies each time frame by a sigmoid-gated combination of median, RMS, and variance; this gated spectrogram is concatenated with the original Log-Mel and the raw-signal Tgram. The network is trained as a self-supervised machine-ID classifier using ArcFace loss, and the negative log probability serves as the anomaly score. On the DCASE 2020 Task 2 development te

Load-bearing premise

The headline claim is load-bearing on the assumption that the quoted baseline numbers are directly comparable to ESTM, even though ESTM was trained on normal sounds from both the development and additional datasets while the baselines were not.

Editorial extensions

If this is right

  • If ESTM's numbers hold, the best published anomalous-sound-detection scores on the DCASE 2020 Task 2 development test set move from transformer/CNN hybrids to a dual-branch selective state-space architecture.
  • The TSG enhancement is parameter-free and consistently improves AUC across all machine types; the full model gains 0.91% average AUC over the same backbone with the un-enhanced input.
  • The time-frequency decoupled model outperforms either single branch by roughly 2 AUC points, supporting the paper's claim that temporal and spectral branches are complementary.
  • The ToyConveyor case shows the largest per-machine improvement, indicating that the method is especially strong on spectrally dominant anomaly patterns.

Reading between the lines

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

  • Inference: The headline margin depends on training ESTM on normal sounds from both the development and additional sets, while the quoted baselines are 'best reported results from their respective papers'; a matched-data retraining is the test that would confirm or erase the 0.84/1.79-point lead.
  • Inference: The TSG gating is parameter-free and computed frame-wise from three statistics, so it could be lifted out of ESTM and applied to other spectrogram-based ASD models as a preprocessing step; the paper only tests it inside its own backbone.
  • Inference: The ablation section's explanation for ToyConveyor—that temporal fusion paradoxically dilutes frequency-domain features—is contradicted by Table II, where the fused STMamba outperforms both single branches on that machine type; a reader should not carry that explanation forward.
  • Inference: On the newer DCASE ASD tasks that add domain shift, ESTM's long-range temporal branch may show different gains; the paper evaluates only the 2020 Task 2 split, so transfer is untested.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes ESTM, a dual-branch (time and frequency) Mamba architecture for anomalous sound detection on the DCASE 2020 Task 2 dataset. Inputs are a concatenation of a Log-Mel spectrogram, a TriStat-Gating (TSG) enhanced spectrogram, and the Tgram raw-audio feature. The two Mamba branches scan time and frequency patches separately and are fused before an ArcFace classification head. The authors report the best average AUC (95.76%) and pAUC (90.49%) in Table I and present ablations (Table II) showing that both the TSG module and the dual-path STMamba contribute to performance.

Significance. If the reported results are protocol-matched and reproducible, the paper would present a strong new application of state-space models to ASD, with a plausible architectural justification and consistent internal ablations. The TSG enhancement idea is simple and potentially reusable. Credit is due for the clear ablation design and for evaluating on a public benchmark. However, the central SOTA claim rests on an uncontrolled comparison with published baselines, the results are single-run point estimates, and several numeric statements in the text do not match the tables. These issues materially limit the strength of the contribution as currently written.

major comments (4)
  1. [§III-A and §III-C, Table I] The SOTA claim is not protocol-matched. Section III-A states that ESTM trains on normal sounds from the development and additional datasets, while Section III-C says the baselines in Table I are 'best reported results from their respective papers.' It is not reported whether those baselines also used the additional training data. If they did not, the reported margins (e.g., 0.84%/1.79% or the actual Table I differences) could be due to extra training data rather than the architecture. Please retrain baselines under the same protocol, or clearly list which baselines used additional data and explicitly restrict the SOTA wording to the matched setting.
  2. [Section III-B, Table I, Table II] All results are reported as single-run point estimates with no seeds, error bars, or significance tests. Since several improvements are small (e.g., the average AUC gain of the TSG module in Table II is about 0.9 percentage points), the reader cannot assess whether these are stable effects or training noise. Please provide multiple runs (at least 3–5 seeds), report mean±std, and ideally a paired test across machine types.
  3. [Abstract, §III-C, §III-D, Table I, Table II] There are multiple numerical inconsistencies. The abstract and §III-C claim an improvement of 0.84% AUC and 1.79% pAUC over ASD-AFPA, but Table I implies 95.76−93.91=1.85 and 90.49−87.95=2.54. In §III-D, the text says STMamba with ESTgram improves average AUC by 0.91% over STgram, but Table II gives 95.76−94.90=0.86; the stated 2.07%/1.94% gains over SMamba/TMamba also do not match the table means (differences are 1.94/1.82). The ToyConveyor gains in §III-C (7.99%/4.50%) do not correspond to the values in Table I either. These numbers need to be recalculated and corrected, and the tables/text must be consistent.
  4. [§III-A, Table I caption] The evaluation split is described ambiguously. Section III-A says 'the test data ... from the development dataset' was used, but Table I's caption reads 'on the test data of the development and additional datasets.' This matters for baseline comparability and for reproducing the results. Please specify exactly which test folds were used (development, additional, or both) and confirm that the quoted baselines used the same folds.
minor comments (5)
  1. [Eq. (1) and §I contributions] The TSG module is described as 'parameter-free,' but Eq. (1) includes a scaling factor α that is manually set to 2 in §III-B. This should be rephrased, e.g., 'with a single scalar hyperparameter,' to avoid contradiction.
  2. [Algorithm 1] In Algorithm 1, a Conv2D operation is applied to the token sequence. Since the input is a 1D token sequence, the distinction between Conv2D and a 1D/2D convolution should be clarified for reproducibility.
  3. [Section II-B, Eq. (4)–(5)] The notation W is used for the projection matrix and also for the frequency window size in the preceding paragraph; please disambiguate.
  4. [Section III-C] The phrase 'state-of-the-art ASD-AFPA method' is odd because ASD-AFPA is a baseline in Table I that is outperformed by TASTgram(NAMix) in average AUC. The text should state the best baseline actually used for comparison.
  5. [General] No code is released. Providing code or detailed hyperparameter/config files would substantially aid reproducibility, especially given the single-run results.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: ESTM's claims rest on benchmark experiments and ablations, not on a derivation that reduces to its inputs.

full rationale

The paper's central claim is empirical: ESTM achieves higher AUC/pAUC than prior methods on the DCASE 2020 Task 2 dataset. No equation or design component is defined in terms of the target metric. The TSG module uses fixed, parameter-free statistics (median, RMS, variance) with a manually set scaling factor alpha=2; patch counts I=12 and J=16 are fixed hyperparameters, not fitted to the test set. The STMamba architecture follows standard Mamba/SSM formulations, and no "prediction" is constructed from a fitted constant. The ablation study (Table II) internally supports the contributions of TSG and the dual-path design. The external comparison in Table I uses 'best reported results from their respective papers,' which raises a protocol-mismatch concern (training on both development and additional data versus baselines trained only on the development set), but this is an experimental-validity issue, not circular reasoning. The paper does not rely on self-citations: references [6], [24], [11], and [9] are works by other author groups, and no load-bearing argument reduces to an unverified self-citation. There is no renaming of a known result as a new derivation. Therefore the circularity score is 0.

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

The central claim is empirical, so no derivation-level free parameters appear. The listed hyperparameters and domain assumptions are standard for a deep-learning benchmark paper; the key additional assumption is the extra training data protocol.

free parameters (5)
  • TSG scaling factor alpha = 2
    Set by hand in Section III-B; controls gating sharpness and affects the enhanced spectrogram.
  • Time patch count I = 12
    Chosen in Section III-B; defines segmentation granularity of the time path.
  • Frequency patch count J = 16
    Chosen in Section III-B; defines segmentation granularity of the frequency path.
  • STFT window/hop and Mel bins = 1024/512/128
    Preprocessing choices in Section III-B that determine input spectrogram shape.
  • Learning rate, batch size, epochs = 1e-4, 128, 200
    Training hyperparameters in Section III-B; no search or sensitivity analysis reported.
assumptions (4)
  • domain assumption DCASE 2020 Task 2 labels and machine IDs are correct and the normal-only training distribution is representative.
    The whole method is trained on normal sounds and evaluated against this public benchmark (Section III-A).
  • domain assumption Machine ID classification with ArcFace loss yields anomaly scores via negative log probability.
    Borrowed from STgram-MFN [6] and ArcFace [26]; no theoretical guarantee that anomalous sounds produce low probability, it is an empirical design choice (Section II-C).
  • domain assumption Selective state-space models capture long-range time-frequency dependencies better than local receptive fields.
    Central architectural premise from Mamba [14]; assumed to transfer to ASD (Sections I, II-B).
  • domain assumption Training on additional-dataset normal sounds does not distort the development-set anomaly detection task.
    The method uses extra training data not used by most baselines; this is an assumption about domain consistency (Section III-A).

how reviews work

0 comments
Cite this review

Pith. "Pith review of ESTM: An Enhanced Dual-Branch Spectral-Temporal Mamba for Anomalous Sound Detection." pith.science (2026). https://pith.science/paper/W5OEQCES

@misc{pith2026250902471,
  author       = {Pith},
  title        = {Pith review of: ESTM: An Enhanced Dual-Branch Spectral-Temporal Mamba for Anomalous Sound Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W5OEQCES}},
  note         = {Machine review of arXiv:2509.02471}
}
read the original abstract

The core challenge in industrial equipment anoma lous sound detection (ASD) lies in modeling the time-frequency coupling characteristics of acoustic features. Existing modeling methods are limited by local receptive fields, making it difficult to capture long-range temporal patterns and cross-band dynamic coupling effects in machine acoustic features. In this paper, we propose a novel framework, ESTM, which is based on a dual-path Mamba architecture with time-frequency decoupled modeling and utilizes Selective State-Space Models (SSM) for long-range sequence modeling. ESTM extracts rich feature representations from different time segments and frequency bands by fusing enhanced Mel spectrograms and raw audio features, while further improving sensitivity to anomalous patterns through the TriStat-Gating (TSG) module. Our experiments demonstrate that ESTM improves anomalous detection performance on the DCASE 2020 Task 2 dataset, further validating the effectiveness of the proposed method.

Figures

Figures reproduced from arXiv: 2509.02471 by the authors.

Figure 1
Figure 1. Framework of the Enhanced Spectral Temporal Mamba(ESTM). [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The t-SNE visualization comparison of STMamba [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 29 canonical work pages

  1. [1]

    Description and discussion on DCASE2020 challenge task2: Unsupervised anomalous sound detection for machine condition monitoring,

    Y . Koizumi et al., “Description and discussion on DCASE2020 challenge task2: Unsupervised anomalous sound detection for machine condition monitoring,” in Proceedings of Detection and Classification of Acoustic Scenes and Events 2020 Workshop (DCASE2020) , November 2020, pp. 81–85

    work page 2020

  2. [2]

    Anomalous sound detection based on interpolation deep neural network,

    K. Suefusa, T. Nishida, H. Purohit, R. Tanabe, T. Endo, and Y . Kawaguchi, “Anomalous sound detection based on interpolation deep neural network,” in Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , Barcelona, Spain, 2020, IEEE, pp. 271–275

    work page 2020

  3. [3]

    Self-supervised classification for detecting anomalous sounds,

    R. Giri, S. Tenneti, F. Cheng, K. Helwani, U. Isik, and A. Krishnaswamy, “Self-supervised classification for detecting anomalous sounds,” in Proceedings of Detection and Classification of Acoustic Scenes and Events 2020 Workshop (DCASE) , 2020, pp. 46–50

    work page 2020

  4. [4]

    ID-Conditioned Auto-Encoder for Unsupervised Anomaly Detection

    Sławomir Kapka, “Id-conditioned auto-encoder for unsupervised anomaly detection,” arXiv preprint arXiv:2007.05314 , 2020

    work page Pith review arXiv 2007

  5. [5]

    Flow- based self-supervised density estimation for anomalous sound detection,

    K. Dohi, T. Endo, H. Purohit, R. Tanabe, and Y . Kawaguchi, “Flow- based self-supervised density estimation for anomalous sound detection,” in Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , Toronto, ON, Canada, 2021, IEEE, pp. 336–340

    work page 2021

  6. [6]

    Anomalous sound detection using spectral-temporal information fusion,

    Y . Liu, J. Guan, Q. Zhu, and W. Wang, “Anomalous sound detection using spectral-temporal information fusion,” in Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, 2022, IEEE, pp. 816–820

    work page 2022

  7. [7]

    Sw-wavenet: learn- ing representation from spectrogram and wavegram using wavenet for anomalous sound detection,

    Haihui Chen, Likai Ran, Xixia Sun, and Chao Cai, “Sw-wavenet: learn- ing representation from spectrogram and wavegram using wavenet for anomalous sound detection,” in ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2023, pp. 1–5

    work page 2023

  8. [8]

    Noise supervised contrastive learning and feature-perturbed for anomalous sound detection,

    Shun Huang, Zhihua Fang, and Liang He, “Noise supervised contrastive learning and feature-perturbed for anomalous sound detection,” in ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2025, pp. 1–5

    work page 2025

Show all 31 references
  1. [9]

    Noisy-arcmix: Additive noisy angular margin loss combined with mixup for anomalous sound de- tection,

    Soonhyeon Choi and Jung-Woo Choi, “Noisy-arcmix: Additive noisy angular margin loss combined with mixup for anomalous sound de- tection,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2024, pp. 516–520

    2024

  2. [10]

    First- shot unsupervised anomalous sound detection with unknown anomalies estimated by metadata-assisted audio generation,

    Hejing Zhang, Qiaoxi Zhu, Jian Guan, Haohe Liu, Feiyang Xiao, Jiantong Tian, Xinhao Mei, Xubo Liu, and Wenwu Wang, “First- shot unsupervised anomalous sound detection with unknown anomalies estimated by metadata-assisted audio generation,” in ICASSP 2024 - 2024 IEEE Internatio...

    2024

  3. [11]

    A dual- path framework with frequency-and-time excited network for anomalous sound detection,

    Yucong Zhang, Juan Liu, Yao Tian, Haifeng Liu, and Ming Li, “A dual- path framework with frequency-and-time excited network for anomalous sound detection,” in ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2024, pp. 1266–1270

    2024

  4. [12]

    Time-weighted frequency domain audio representation with gmm estimator for anomalous sound detection,

    Jian Guan, Youde Liu, Qiaoxi Zhu, Tieran Zheng, Jiqing Han, and Wenwu Wang, “Time-weighted frequency domain audio representation with gmm estimator for anomalous sound detection,” in ICASSP 2023- 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (IC...

    2023

  5. [13]

    Cst-former: Transformer with channel-spectro-temporal attention for sound event localization and detection,

    Yusun Shul and Jung-Woo Choi, “Cst-former: Transformer with channel-spectro-temporal attention for sound event localization and detection,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2024, pp. 8686– 8690

    2024

  6. [14]

    Mamba: Linear-time sequence modeling with selective state spaces,

    Albert Gu and Tri Dao, “Mamba: Linear-time sequence modeling with selective state spaces,” arXiv preprint arXiv:2312.00752 , 2023

    2023 arXiv

  7. [15]

    Transformers are ssms: generalized models and efficient algorithms through structured state space duality,

    Tri Dao and Albert Gu, “Transformers are ssms: generalized models and efficient algorithms through structured state space duality,” in Proceedings of the 41st International Conference on Machine Learning . 2024, ICML’24, JMLR.org

    2024

  8. [16]

    Selective structured state-spaces for long- form video understanding,

    Jue Wang, Wentao Zhu, Pichao Wang, Xiang Yu, Linda Liu, Mohamed Omar, and Raffay Hamid, “Selective structured state-spaces for long- form video understanding,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023, pp. 6387–6397

    2023

  9. [17]

    Tf-mamba: A time-frequency net- work for sound source localization,

    Yang Xiao and Rohan Kumar Das, “Tf-mamba: A time-frequency net- work for sound source localization,” arXiv preprint arXiv:2409.05034 , 2024

    2024 arXiv

  10. [18]

    Audio mamba: Bidirectional state space model for audio representation learning,

    Mehmet Hamza Erol, Arda Senocak, Jiu Feng, and Joon Son Chung, “Audio mamba: Bidirectional state space model for audio representation learning,” IEEE Signal Processing Letters, vol. 31, pp. 2975–2979, 2024

    2024

  11. [19]

    Ssamba: Self-supervised audio representation learning with mamba state space model,

    Siavash Shams, Sukru Samet Dindar, Xilin Jiang, and Nima Mesgarani, “Ssamba: Self-supervised audio representation learning with mamba state space model,” in 2024 IEEE Spoken Language Technology Workshop (SLT), 2024, pp. 1053–1059

    2024

  12. [20]

    Mamba in speech: Towards an alternative to self-attention,

    Xiangyu Zhang, Qiquan Zhang, Hexin Liu, Tianyi Xiao, Xinyuan Qian, Beena Ahmed, Eliathamby Ambikairajah, Haizhou Li, and Julien Epps, “Mamba in speech: Towards an alternative to self-attention,” IEEE Transactions on Audio, Speech and Language Processing , vol. 33, pp. 1933–1948, 2025

    1933

  13. [21]

    Pooling methods in deep neural networks, a review,

    Hossein Gholamalinezhad and Hossein Khosravi, “Pooling methods in deep neural networks, a review,”arXiv preprint arXiv:2009.07485, 2020

    2009 arXiv

  14. [22]

    Root mean square layer normalization,

    Biao Zhang and Rico Sennrich, “Root mean square layer normalization,” Advances in Neural Information Processing Systems , vol. 32, 2019

    2019

  15. [23]

    Spectral conversion based on maximum likelihood estimation considering global variance of converted parameter,

    Tomoki Toda, Alan W Black, and Keiichi Tokuda, “Spectral conversion based on maximum likelihood estimation considering global variance of converted parameter,” in Proceedings.(ICASSP’05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005. IEEE, 20...

    2005

  16. [24]

    Anomalous sound detection using self-attention-based frequency pattern analysis of machine sounds,

    Hejing Zhang, Jian Guan, Qiaoxi Zhu, Feiyang Xiao, and Youde Liu, “Anomalous sound detection using self-attention-based frequency pattern analysis of machine sounds,” in Interspeech 2023 , 2023, pp. 336–340

    2023

  17. [25]

    Vision mamba: Efficient visual representation learning with bidirectional state space model,

    Lianghui Zhu, Bencheng Liao, Qian Zhang, Xinlong Wang, Wenyu Liu, and Xinggang Wang, “Vision mamba: Efficient visual representation learning with bidirectional state space model,” in Proceedings of the 41st International Conference on Machine Learning , Ruslan Salakhutdinov, Z...

    2024

  18. [26]

    Arcface: Additive angular margin loss for deep face recognition,

    J. Deng, J. Guo, N. Xue, and S. Zafeiriou, “Arcface: Additive angular margin loss for deep face recognition,” in Proceedings of IEEE/CVF conference on computer vision and pattern recognition (CVPR) , Long Beach, CA, USA, 2019, pp. 4690–4699

    2019

  19. [27]

    MIMII Dataset: Sound dataset for malfunctioning industrial machine investigation and inspection,

    Harsh Purohit, Ryo Tanabe, Takeshi Ichige, Takashi Endo, Yuki Nikaido, Kaori Suefusa, and Yohei Kawaguchi, “MIMII Dataset: Sound dataset for malfunctioning industrial machine investigation and inspection,” in Proceedings of the Detection and Classification of Acoustic Scenes a...

    2019

  20. [28]

    ToyADMOS: A dataset of miniature-machine operating sounds for anomalous sound detection,

    Yuma Koizumi, Shoichiro Saito, Hisashi Uematsu, Noboru Harada, and Keisuke Imoto, “ToyADMOS: A dataset of miniature-machine operating sounds for anomalous sound detection,” in Proceedings of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), No...

    2019

  21. [29]

    Sigmoid-weighted linear units for neural network function approximation in reinforcement learning,

    Stefan Elfwing, Eiji Uchibe, and Kenji Doya, “Sigmoid-weighted linear units for neural network function approximation in reinforcement learning,” Neural networks, vol. 107, pp. 3–11, 2018

    2018

  22. [30]

    Decoupled weight decay regulariza- tion,

    Ilya Loshchilov and Frank Hutter, “Decoupled weight decay regulariza- tion,” in International Conference on Learning Representations , 2017

    2017

  23. [31]

    t-distributed stochastic neighbor em- bedding (t-sne): A tool for eco-physiological transcriptomic analysis,

    Matthew C Cieslak, Ann M Castelfranco, Vittoria Roncalli, Petra H Lenz, and Daniel K Hartline, “t-distributed stochastic neighbor em- bedding (t-sne): A tool for eco-physiological transcriptomic analysis,” Marine genomics, vol. 51, pp. 100723, 2020

    2020

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.