Pith. sign in

REVIEW 4 major objections 5 minor 47 references

Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read SCLIFD claims a three-part framework—contrastive knowledge distillation, boundary-sample replay, and a balanced random forest—keeps class-incremental fault diagnosis accurate when fault data are scarce, imbalanced, and long-tailed.

desk verdict Solid engineering framework for class-incremental fault diagnosis with limited data, but the empirical claims need variance reporting and a validation-based memory-size selection before they should be taken as established. read the letter →

arxiv 2501.09525 v2 pith:PHJKGZ7X submitted 2025-01-16 cs.LG cs.AI

Hanrong Zhang , Yifei Yao , Zixuan Wang , Jiayuan Su , Mengxuan Li , Peng Peng , Hongwei Wang This is my paper · ORCID
classification cs.LGcs.AI
keywords classincrementallearningfaultdiagnosissupervisedcontrastiveknowledgedistillationcatastrophicforgettingimbalancelong-taileddistributionexemplarreplay
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 tries to establish that class-incremental fault diagnosis can work even when new fault classes arrive with very few samples and the overall data are imbalanced or long-tailed. It proposes SCLIFD, a framework that combines supervised contrastive learning with feature-space knowledge distillation, plus a boundary-focused exemplar replay method and a balanced random forest classifier. The claimed payoff is that a fault-diagnosis model can keep learning new faults over successive sessions without catastrophic forgetting and without biasing predictions toward the normal class. If true, industrial diagnosis systems could be updated periodically with new fault types instead of being retrained from scratch.

What carries the argument

The central machinery is the Supervised Contrastive Knowledge Distillation (SCKD) loss, which adds a cross-session feature-space distillation term to a supervised contrastive loss: the frozen previous-session extractor serves as teacher, the current extractor as student, and the student must reproduce the teacher's pairwise similarity structure over old features. This is paired with Marginal Exemplar Selection (MES), which stores samples farthest from the class mean in feature space, and a Balanced Random Forest (BRF) classifier that fits each tree to equal numbers of minority and majority exemplars. The three components are designed to work as a chain: SCKD produces the clean feature space that lets MES find boundary samples, MES fills the replay buffer with informative samples, and BRF makes the final class decision on the embedded, rebalanced exemplars.

What would settle it

Train SCLIFD on a fault type whose diagnosis depends on the order of events inside the signal, with and without the segment-shuffle augmentation; if the shuffled model's accuracy on that fault falls clearly below the unshuffled model's, the augmentation is creating wrong positive pairs and the representation stage is unreliable.

Watch

Extended reading notes

Core claim

The central claim is that SCLIFD, built from three coordinated components, outperforms eight existing class-incremental learning methods on imbalanced and long-tailed fault diagnosis. The supervised contrastive knowledge distillation loss lets the current session's feature extractor learn discriminative features while emulating the previous session's feature space, reducing forgetting. The Marginal Exemplar Selection method replays the hardest boundary samples rather than centroid samples, keeping the decision boundaries visible to later sessions. The Balanced Random Forest classifier counters normal-class bias by training each tree on balanced bootstraps. The paper reports average accuracies of 90.23% and 84.27% on the imbalanced and long-tailed Tennessee Eastman Process cases, and 99.73% and 99.44% on the two Multiphase Flow Facility long-tailed cases, all above the compared baselines in the reported tables.

Load-bearing premise

The whole method assumes that the data augmentation used during training—cutting a random segment out of each fault signal, shuffling it, and putting it back—does not change the fault's identity, so the contrastive loss never pairs two different faults as if they were the same.

Editorial extensions

If this is right

  • If the reported numbers hold, industrial systems can add new fault classes through periodic offline updates using a small constant-sized memory buffer rather than full retraining.
  • The gap at the final incremental session (e.g., 72.25% versus 54.02% for iCaRL on imbalanced TEP) implies that feature-space distillation plus boundary-sample replay is a workable recipe against catastrophic forgetting under limited fault data.
  • The per-component ablation results imply that each of the three modules—SCKD, MES, and BRF—contributes distinct value, so the framework does not rest on a single trick.
  • The generalization experiment across different TEP fault selections, with an average accuracy of 85.86%, suggests the method is not tuned to one particular set of fault signatures.
  • The sensitivity analysis suggests that small memory buffers (K=100 or lower) can sustain accuracy, which matters for deployment with strict memory constraints.

Reading between the lines

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

  • The paper leaves implicit that the segment-shuffle augmentation used for contrastive learning is the least protected assumption: if a fault signature depends on the temporal order inside the shuffled segment, the augmentation can create contradicting positive pairs and corrupt the learned features.
  • Because MES selects exemplars using the current feature extractor, the 'marginal' set is a moving target; a testable extension is to re-score exemplars after each session and refresh the buffer so boundary samples remain boundary samples.
  • The Balanced Random Forest is trained only on the small exemplar buffer, so with very small memories the balanced bootstrap may become degenerate; mixing synthetic oversampling into the classifier stage is a natural extension.
  • The three-component recipe is not obviously limited to fault diagnosis and could be tested in other time-series class-incremental domains with rare classes and a normal-bias problem, such as predictive maintenance or anomaly detection in process control.
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. The paper proposes SCLIFD, a class-incremental fault diagnosis framework for limited, imbalanced, and long-tailed fault data. It combines supervised contrastive knowledge distillation (SCKD) for representation learning, a marginal exemplar selection (MES) strategy for replay memory, and a balanced random forest (BRF) classifier to counter class imbalance. The method is evaluated on the Tennessee Eastman Process (TEP) and Multiphase Flow Facility (MFF) datasets in one imbalanced and three long-tailed settings, and compared against six baselines. The authors report consistently higher average accuracies, e.g., 90.23% and 84.27% on TEP imbalanced and long-tailed, and 99.73% and 99.44% on MFF long-tailed 1 and 2, and provide ablations for each component.

Significance. If the reported results are reproducible and robust, SCLIFD addresses a practically important gap: most class-incremental fault diagnosis work assumes balanced data, whereas real fault data are typically scarce and imbalanced. The design is plausible, the three components are clearly motivated, and the ablation study gives some support for each contribution. The paper also provides code. However, the central claim of superiority over all baselines is an empirical claim, and the current experimental protocol does not establish it statistically: all numbers are single-run, the memory size K is selected on what appears to be the test set, and one of the two datasets uses only two exemplars per class for classifier training. These issues are load-bearing because the ranking across methods is exactly what the paper promises.

major comments (4)
  1. [§IV-B, Tables III–IV] Every accuracy in Tables III and IV comes from a single run, with no standard deviations, no number of seeds, and no repeated class orders. The headline comparison includes large gaps, e.g., TEP imbalanced session 5 (72.25% vs. 54.02% for iCaRL), but some differences are small, e.g., MFF Long-Tailed 1: 97.94% (Mixed) vs. 99.73% (full SCLIFD) in Table IV. Without variance estimates it is impossible to tell whether the claimed ordering is stable under random initialization, data order, or class order. Please report means and standard deviations over at least five seeds (and preferably multiple class orders) for the main comparisons and the key ablation rows.
  2. [§IV-E, Fig. 8] The memory buffer size K is selected in a sensitivity analysis by looking at average accuracy, and the same average accuracy is the reported evaluation metric in Table III. This amounts to test-set model selection for SCLIFD if the curves in Fig. 8 are computed on the test set, and the paper does not state whether the baselines were given the same per-dataset K tuning or were evaluated with a single fixed K. If baselines were not tuned in the same way, the comparison is biased. Please either select K on a validation split, or report accuracy as a function of K for all methods and adopt an identical selection protocol.
  3. [§IV-A, Table I and Algorithm 3] For MFF the memory size is K=10 and the number of classes is 5, so Algorithm 1 (line 3) allocates m=2 exemplars per class. Algorithm 3 then trains the balanced random forest on features extracted from these two exemplars per class. This is an extremely small training set, and it is not clear whether the high MFF accuracies reflect the method's components or the triviality of the classifier problem under this construction. Please report how accuracy varies with m for MFF, clarify the exact per-class exemplar count, and discuss the stability of the BRF when trained on two samples per class.
  4. [§IV-A.3, data augmentation] The supervised contrastive representation learning uses a random segment-shuffle augmentation that reorders a segment of each input and integrates it back into the sequence. For fault signatures that depend on temporal order, such as step changes, ramps, or valve sticking, this augmentation may not preserve the fault class, creating contradictory positive pairs in the SCL loss. The paper does not validate that the augmentation preserves fault identity for the TEP and MFF fault types used. Please add an experiment comparing this augmentation with a label-preserving augmentation or a quantitative test of augmentation-induced invariance, since the entire representation-learning stage relies on this assumption.
minor comments (5)
  1. [Algorithm 2, line 4] The expression for Lencoder contains an outer sum over (xi, yi) followed by sums over y=1..t and y=1..s-1, which appears to multiply the per-batch losses by the batch size and by the number of classes. Please clarify the intended indexing and whether Lscl and Ldis are per-sample or per-batch quantities.
  2. [§IV-B.2] The text refers to 'MFF's imbalanced case', but Table I only defines MFF as long-tailed 1 and long-tailed 2; please correct this inconsistency.
  3. [Abstract and Table I] There are small textual slips: 'distiLlation' in the abstract, 'Dateset' in Table I, and the table column layout for 'Novel-Class Shot' and 'Training Set' is ambiguous and should be reformatted.
  4. [§III-B, Eq. (4)] In Eq. (4), P(zi;za) is defined as a softmax over all other samples, so Ldis is a cross-entropy between pairwise similarity distributions. Please state explicitly whether the teacher and student use the same negative sets and whether gradients flow through the student's negative samples; this is important for reproducibility.
  5. [References] Reference [19] is an earlier arXiv paper with the same name and likely substantial overlap with this submission. The relationship between the two works should be stated explicitly so that readers can assess the novelty of the journal version.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported gains are empirical comparisons; the framework components are defined independently and the self-citations are not load-bearing.

full rationale

I walked the paper's claimed derivation chain. There is no equation-level reduction: SCKD is defined by Eqs. (2)-(4), MES by Eqs. (5)-(6), and the BRF classifier by the balanced sampling procedure in Sec. II-C; none of these definitions contains the test accuracies they are claimed to produce. The nearest candidate, memory-size selection in Sec. IV-E, is a hyperparameter sweep (K chosen from average-accuracy curves), not a fitted prediction; the same K is then used for all methods (Tab. I), so SCLIFD's advantage is not forced by construction. Self-citations appear in the introduction ([19] for 'growing interest'), in a limitation motivation ([36]), and in dataset-setting choices ([39]); none of these is the load-bearing argument for the central 'outperforms' claim, which rests on the Tables III-V comparisons against externally defined baselines (iCaRL, EEIL, BiC, SAVC, WaRP-CIFSL, BiDistFSCIL). The single-run/test-set-tuning concerns are reproducibility risks, not circularity, under the hard rules. I therefore report no circular steps.

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

The paper introduces no new physical or mathematical entities. It relies on standard supervised contrastive learning, feature-space distillation, exemplar replay, and balanced random forests, plus a small number of hand-chosen hyperparameters. The most fragile assumptions are the validity of the augmentation and the representativeness of tiny exemplar buffers.

free parameters (4)
  • Temperature tau = 0.07
    Used in both the supervised contrastive loss and the distillation similarity distributions; fixed by hand without a reported sweep.
  • Learning rate = 0.01
    Adam optimizer learning rate for the feature extractor; no schedule or decay is reported.
  • Memory buffer size K = 100 for TEP imbalanced, 40 for TEP long-tailed, 10 and 5 for MFF long-tailed cases
    Selected from test-set accuracy curves in Fig. 8, so the reported performance is tuned to the test set rather than a held-out validation set.
  • Balanced random forest ensemble size = not reported
    The number of trees N in the balanced random forest is never stated, although the method and its variance depend on it.
assumptions (3)
  • domain assumption Random segment-shuffle augmentation preserves fault-relevant information, so positive pairs remain valid for supervised contrastive learning.
    Used in the SCL training described in Sec. IV-A (parameter setting); if segment ordering is discriminative for a fault, this augmentation would create false positive pairs.
  • domain assumption Feature distillation from the previous session's frozen feature extractor transfers old-class knowledge without requiring access to old data beyond the exemplar buffers.
    Foundation of SCKD in Eq. 4 and Algorithm 2; no proof is given that similarity-distribution matching is sufficient to prevent forgetting.
  • domain assumption Marginal samples, those farthest from the class centroid in feature space, are the most valuable exemplars for preventing forgetting.
    MES criterion in Eq. 6, motivated by Fig. 3; the paper provides no theoretical or controlled empirical justification that boundary samples are optimal.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation." pith.science (2026). https://pith.science/paper/PHJKGZ7X

@misc{pith2026250109525,
  author       = {Pith},
  title        = {Pith review of: Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PHJKGZ7X}},
  note         = {Machine review of arXiv:2501.09525}
}
read the original abstract

Class-incremental fault diagnosis requires a model to adapt to new fault classes while retaining previous knowledge. However, limited research exists for imbalanced and long-tailed data. Extracting discriminative features from few-shot fault data is challenging, and adding new fault classes often demands costly model retraining. Moreover, incremental training of existing methods risks catastrophic forgetting, and severe class imbalance can bias the model's decisions toward normal classes. To tackle these issues, we introduce a Supervised Contrastive knowledge distiLlation for class Incremental Fault Diagnosis (SCLIFD) framework proposing supervised contrastive knowledge distillation for improved representation learning capability and less forgetting, a novel prioritized exemplar selection method for sample replay to alleviate catastrophic forgetting, and the Random Forest Classifier to address the class imbalance. Extensive experimentation on simulated and real-world industrial datasets across various imbalance ratios demonstrates the superiority of SCLIFD over existing approaches. Our code can be found at https://github.com/Zhang-Henry/SCLIFD_TII.

Figures

Figures reproduced from arXiv: 2501.09525 by the authors.

Figure 1
Figure 1. Class incremental learning for fault diagnosis under limited fault data. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. iCaRL’s T-SNE visual￾ization under the imbalanced case of TEP dataset. iCaRL cannot ex￾tract discriminative features from imbalanced classes. Class 0 Center of Class 0 Class 1 Center of Class 1 Class 2 Center of Class 2 Class 3 Center of Class 3 Misclassified [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 4
Figure 4. The general framework of the proposed method SCLIFD for fault diagnosis under limited fault data, which is detailed explained in Sec. III-A. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Experiment result comparisons of different methods on TEP and MFF Dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: T-SNE visualization of each class’s features extracted by SCLIFD. It extracts discriminative features for different classes over multiple sessions. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The node network graph for features extracted by our SCLIFD [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Average Accuracies of different K values under TE and MFF datasets. [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 45 canonical work pages

  1. [19]

    SCLIFD:Supervised Contrastive Knowledge Distillation for Incremental Fault Diagnosis under Limited Fault Data

    P. Peng, H. Zhang, M. Li, G. Peng, H. Wang, and W. Shen, “Sclifd:supervised contrastive knowledge distillation for incremental fault diagnosis under limited fault data,” 2023. [Online]. Available: https://arxiv.org/abs/2302.05929

    work page Pith review arXiv 2023

  2. [1]

    Gaussian mixture variational-based transformer domain adaptation fault diagnosis method and its application in bearing fault diagnosis,

    Y . An, K. Zhang, Y . Chai, Z. Zhu, and Q. Liu, “Gaussian mixture variational-based transformer domain adaptation fault diagnosis method and its application in bearing fault diagnosis,” IEEE Transactions on Industrial Informatics, vol. 20, no. 1, pp. 615–625, 2024

    work page 2024

  3. [2]

    A hierarchical training-convolutional neural network for imbalanced fault diagnosis in complex equipment,

    Y . Gao, L. Gao, X. Li, and S. Cao, “A hierarchical training-convolutional neural network for imbalanced fault diagnosis in complex equipment,” IEEE Transactions on Industrial Informatics , vol. 18, no. 11, pp. 8138– 8145, 2022

    work page 2022

  4. [3]

    An order-invariant and interpretable dilated convolution neural network for chemical process fault detection and diagnosis,

    M. Li, P. Peng, H. Sun, M. Wang, and H. Wang, “An order-invariant and interpretable dilated convolution neural network for chemical process fault detection and diagnosis,”IEEE Transactions on Automation Science and Engineering, 2023

    work page 2023

  5. [4]

    Hard Sample Mining Enabled Supervised Contrastive Feature Learning for Wind Turbine Pitch System Fault Diagnosis

    Z. Wang, B. Qin, M. Li, M. D. Butala, H. Wang, P. Peng, and H. Wang, “Hard sample mining enabled contrastive feature learning for wind turbine pitch system fault diagnosis,” arXiv preprint arXiv:2306.14701 , 2023

    work page Pith review arXiv 2023

  6. [5]

    Generalized out-of-distribution fault diagnosis (goofd) via internal con- trastive learning,

    X. Wang, H. Zhang, X. Qiao, K. Ma, S. Tao, P. Peng, and H. Wang, “Generalized out-of-distribution fault diagnosis (goofd) via internal con- trastive learning,” IEEE Transactions on Industrial Informatics , vol. 20, no. 8, pp. 9987–9996, 2024

    work page 2024

  7. [6]

    Imbalanced chemical process fault diagnosis using balancing gan with active sample selection,

    P. Peng, H. Zhang, X. Wang, W. Huang, and H. Wang, “Imbalanced chemical process fault diagnosis using balancing gan with active sample selection,” IEEE Sensors Journal , vol. 23, no. 13, pp. 14 826–14 833, 2023

    work page 2023

  8. [7]

    Multi-gate mixture-of- expert combined with synthetic minority over-sampling technique for multimode imbalanced fault diagnosis,

    W. Huang, H. Zhang, P. Peng, and H. Wang, “Multi-gate mixture-of- expert combined with synthetic minority over-sampling technique for multimode imbalanced fault diagnosis,” in 2023 26th International Con- ference on Computer Supported Cooperative Work in Design (CSCWD) , 2023, pp. 456–461

    work page 2023

Show all 47 references
  1. [8]

    Sccam: Supervised contrastive convolutional attention mechanism for ante-hoc interpretable fault diagnosis with limited fault samples,

    M. Li, P. Peng, J. Zhang, H. Wang, and W. Shen, “Sccam: Supervised contrastive convolutional attention mechanism for ante-hoc interpretable fault diagnosis with limited fault samples,” IEEE Transactions on Neural Networks and Learning Systems , 2023

    2023

  2. [9]

    An imbalanced semi-supervised wind turbine blade icing detection method based on contrastive learning,

    Z. Wang, B. Qin, H. Sun, J. Zhang, M. D. Butala, C. Demartino, P. Peng, and H. Wang, “An imbalanced semi-supervised wind turbine blade icing detection method based on contrastive learning,” Renewable Energy, vol. 212, pp. 251–262, 2023

    2023

  3. [10]

    Imbalance fault diagnosis under long-tailed distribution: Challenges, solutions and prospects,

    Z. Chen, J. Chen, Y . Feng, S. Liu, T. Zhang, K. Zhang, and W. Xiao, “Imbalance fault diagnosis under long-tailed distribution: Challenges, solutions and prospects,” Knowledge-Based Systems, vol. 258, Dec 2022

    2022

  4. [11]

    Feature selection with imbalanced data for software defect prediction,

    T. M. Khoshgoftaar and K. Gao, “Feature selection with imbalanced data for software defect prediction,” in 2009 International Conference on Machine Learning and Applications , Dec 2009, pp. 235–240

    2009

  5. [12]

    A svm framework for fault detection of the braking system in a high speed train,

    J. Liu, Y .-F. Li, and E. Zio, “A svm framework for fault detection of the braking system in a high speed train,” Mechanical Systems and Signal Processing, vol. 87, pp. 401–409, Mar 2017

    2017

  6. [13]

    Generative adversarial networks,

    I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y . Bengio, “Generative adversarial networks,” Communications of the ACM , vol. 63, no. 11, p. 139–144, Oct 2020

    2020

  7. [14]

    A unified framework incorporating predictive generative denoising autoencoder and deep coral network for rolling bearing fault diagnosis with unbalanced data,

    X. Li, H. Jiang, S. Liu, J. Zhang, and J. Xu, “A unified framework incorporating predictive generative denoising autoencoder and deep coral network for rolling bearing fault diagnosis with unbalanced data,” Measurement, vol. 178, Jun 2021

    2021

  8. [15]

    Transfer fault diagnosis of bearing installed in different machines using enhanced deep auto-encoder,

    H. Zhiyi, S. Haidong, J. Lin, C. Junsheng, and Y . Yu, “Transfer fault diagnosis of bearing installed in different machines using enhanced deep auto-encoder,” Measurement, vol. 152, Feb 2020

    2020

  9. [16]

    Feature space augmentation for long-tailed data,

    P. Chu, X. Bian, S. Liu, and H. Ling, “Feature space augmentation for long-tailed data,” in European Conference on Computer Vision . Springer, 2020, pp. 694–710

    2020

  10. [17]

    McCloskey and N

    M. McCloskey and N. J. Cohen, Catastrophic Interference in Connec- tionist Networks: The Sequential Learning Problem . Academic Press, Jan 1989, vol. 24, pp. 109–165

    1989

  11. [18]

    From uncertainty to clarity: Uncertainty-guided class-incremental learning for limited biomedical samples via semantic expansion,

    Y . Yao and H. Zhang, “From uncertainty to clarity: Uncertainty-guided class-incremental learning for limited biomedical samples via semantic expansion,” 2024. [Online]. Available: https://arxiv.org/abs/2409.07757

    2024 arXiv

  12. [20]

    Cross- domain class incremental broad network for continuous diagnosis of rotating machinery faults under variable operating conditions,

    M. Shi, C. Ding, S. Chang, C. Shen, W. Huang, and Z. Zhu, “Cross- domain class incremental broad network for continuous diagnosis of rotating machinery faults under variable operating conditions,” IEEE Transactions on Industrial Informatics , vol. 20, no. 4, pp. 6356–6368, Apr. 2024

    2024

  13. [21]

    A data-driven time-series fault prediction framework for dynamically evolving large- scale data streaming systems,

    M. Hell, E. Pestana de Aguiar, N. Soares, and L. Goliatt, “A data-driven time-series fault prediction framework for dynamically evolving large- scale data streaming systems,” International Journal of Fuzzy Systems , vol. 24, no. 6, pp. 2831–2844, Sep 2022

    2022

  14. [22]

    Bearing fault diagnosis via incremental learning based on the repeated replay using memory indexing (r-remind) method,

    J. Zheng, H. Xiong, Y . Zhang, K. Su, and Z. Hu, “Bearing fault diagnosis via incremental learning based on the repeated replay using memory indexing (r-remind) method,” Machines, vol. 10, no. 55, May 2022

    2022

  15. [23]

    Broad convolutional neural network based indus- trial process fault diagnosis with incremental learning capability,

    W. Yu and C. Zhao, “Broad convolutional neural network based indus- trial process fault diagnosis with incremental learning capability,” IEEE Transactions on Industrial Electronics , vol. 67, no. 6, pp. 5081–5091, Jun 2020

    2020

  16. [24]

    An imbalance modified con- volutional neural network with incremental learning for chemical fault diagnosis,

    X. Gu, Y . Zhao, G. Yang, and L. Li, “An imbalance modified con- volutional neural network with incremental learning for chemical fault diagnosis,” IEEE Transactions on Industrial Informatics , vol. 18, no. 6, pp. 3630–3639, Jun. 2022

    2022

  17. [25]

    Hsell-net: A heterogeneous sample enhancement network with lifelong learning under industrial small samples,

    Y . Ren, J. Liu, Q. Wang, and H. Zhang, “Hsell-net: A heterogeneous sample enhancement network with lifelong learning under industrial small samples,” IEEE Transactions on Cybernetics , pp. 1–13, 2022

    2022

  18. [26]

    Intelligent fault diagnosis of rotor-bearing system under varying working conditions with modified transfer convolutional neural network and thermal images,

    H. Shao, M. Xia, G. Han, Y . Zhang, and J. Wan, “Intelligent fault diagnosis of rotor-bearing system under varying working conditions with modified transfer convolutional neural network and thermal images,” IEEE Transactions on Industrial Informatics , vol. 17, no. 5, pp. 3488...

    2021

  19. [27]

    Open-set domain adaptation in machinery fault diagnostics using instance-level weighted adversarial learning,

    W. Zhang, X. Li, H. Ma, Z. Luo, and X. Li, “Open-set domain adaptation in machinery fault diagnostics using instance-level weighted adversarial learning,” IEEE Transactions on Industrial Informatics , vol. 17, no. 11, pp. 7445–7455, 2021. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 11

    2021

  20. [28]

    Deep adversarial subdomain adaptation network for intelligent fault diagnosis,

    Y . Liu, Y . Wang, T. W. S. Chow, and B. Li, “Deep adversarial subdomain adaptation network for intelligent fault diagnosis,” IEEE Transactions on Industrial Informatics, vol. 18, no. 9, pp. 6038–6046, 2022

    2022

  21. [29]

    Variable-wise stacked temporal autoencoder for intelligent fault diagnosis of industrial systems,

    L. Liu, Y . Zheng, and S. Liang, “Variable-wise stacked temporal autoencoder for intelligent fault diagnosis of industrial systems,” IEEE Transactions on Industrial Informatics , vol. 20, no. 5, pp. 7545–7555, 2024

    2024

  22. [30]

    Multiscale channel attention- driven graph dynamic fusion learning method for robust fault diagnosis,

    X. Zhang, J. Liu, X. Zhang, and Y . Lu, “Multiscale channel attention- driven graph dynamic fusion learning method for robust fault diagnosis,” IEEE Transactions on Industrial Informatics, vol. 20, pp. 11 002–11 013, 2024

    2024

  23. [31]

    Supervised contrastive learn- ing,

    P. Khosla, P. Teterwak, C. Wang, A. Sarna, Y . Tian, P. Isola, A. Maschinot, C. Liu, and D. Krishnan, “Supervised contrastive learn- ing,” in Advances in Neural Information Processing Systems , vol. 33. Curran Associates, Inc., 2020, pp. 18 661–18 673

    2020

  24. [32]

    Using random forest to learn imbalanced data,

    C. Chen, A. Liaw, and L. Breiman, “Using random forest to learn imbalanced data,” Univ. California, Berkeley, Berkeley, CA, USA, Tech. Rep., 2004

    2004

  25. [33]

    iCaRL: Incremental classifier and representation learning,

    S.-A. Rebuffi, A. Kolesnikov, G. Sperl, and C. H. Lampert, “iCaRL: Incremental classifier and representation learning,” in 2017 IEEE Con- ference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 5533–5542

    2017

  26. [34]

    Herding dynamical weights to learn,

    M. Welling, “Herding dynamical weights to learn,” in Proceedings of the 26th Annual International Conference on Machine Learning , 2009, pp. 1121–1128

    2009

  27. [35]

    End-to-end incremental learning,

    F. M. Castro, M. J. Marin-Jimenez, N. Guil, C. Schmid, and K. Alahari, “End-to-end incremental learning,” in Proceedings of the European Conference on Computer Vision (ECCV) , Sept 2018

    2018

  28. [36]

    Sccam: Supervised contrastive convolutional attention mechanism for ante-hoc interpretable fault diagnosis with limited fault samples,

    M. Li, P. Peng, J. Zhang, H. Wang, and W. Shen, “Sccam: Supervised contrastive convolutional attention mechanism for ante-hoc interpretable fault diagnosis with limited fault samples,” IEEE Transactions on Neural Networks and Learning Systems , pp. 1–12, 2024

    2024

  29. [37]

    Breiman, Classification and regression trees

    L. Breiman, Classification and regression trees . Routledge, 2017

    2017

  30. [38]

    Large scale incremental learning,

    Y . Wu, Y . Chen, L. Wang, Y . Ye, Z. Liu, Y . Guo, and Y . Fu, “Large scale incremental learning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , June 2019

    2019

  31. [39]

    Progressively balanced supervised contrastive representation learning for long-tailed fault diagnosis,

    P. Peng, J. Lu, S. Tao, K. Ma, Y . Zhang, H. Wang, and H. Zhang, “Progressively balanced supervised contrastive representation learning for long-tailed fault diagnosis,” IEEE Transactions on Instrumentation and Measurement, vol. 71, pp. 1–12, 2022

    2022

  32. [40]

    Knowledge distillation: A survey,

    J. Gou, B. Yu, S. J. Maybank, and D. Tao, “Knowledge distillation: A survey,” International Journal of Computer Vision , vol. 129, no. 6, pp. 1789–1819, Jun. 2021

    2021

  33. [41]

    Decentralized control of the tennessee eastman challenge process,

    N. Lawrence Ricker, “Decentralized control of the tennessee eastman challenge process,” Journal of Process Control , vol. 6, no. 4, pp. 205– 221, 1996

    1996

  34. [42]

    Statistical process monitoring of a multiphase flow facility,

    C. Ruiz-C ´arcel, Y . Cao, D. Mba, L. Lao, and R. Samuel, “Statistical process monitoring of a multiphase flow facility,” Control Engineering Practice, vol. 42, pp. 74–88, 2015

    2015

  35. [43]

    Learning without forgetting,

    Z. Li and D. Hoiem, “Learning without forgetting,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 40, pp. 2935–2947, Dec 2018

    2018

  36. [44]

    Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class-incremental learning,

    Z. Song, Y . Zhao, Y . Shi, P. Peng, L. Yuan, and Y . Tian, “Learning with fantasy: Semantic-aware virtual contrastive constraint for few-shot class-incremental learning,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , 2023, pp. 24 183–24 192

    2023

  37. [45]

    Warping the space: Weight space rotation for class-incremental few-shot learning,

    D.-Y . Kim, D.-J. Han, J. Seo, and J. Moon, “Warping the space: Weight space rotation for class-incremental few-shot learning,” in The Eleventh International Conference on Learning Representations , 2023

    2023

  38. [46]

    Few- shot class-incremental learning via class-aware bilateral distillation,

    L. Zhao, J. Lu, Y . Xu, Z. Cheng, D. Guo, Y . Niu, and X. Fang, “Few- shot class-incremental learning via class-aware bilateral distillation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2023, pp. 11 838–11 847. Hanrong Zhang (https://...

    2023

  39. [2022]

    Her research interest is deep learning

    She is currently pursuing a Master’s degree at Zhejiang University, China. Her research interest is deep learning. Zixuan Wang is currently a doctoral student at the College of Biomedical Engineering and Instrument Science in Zhejiang University, Hangzhou, China. He received h...

    2020

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.