Pith. sign in

REVIEW 4 major objections 7 minor 36 references

Explainable Artificial Intelligence based Soft Evaluation Indicator for Arc Fault Diagnosis

T0 review · 4 major / 7 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper proposes XSEI, a soft evaluation indicator that measures whether an arc fault diagnosis model's explanations overlap with a defined physical ground truth of arc faults, arguing that high accuracy alone does not certify that a…

desk verdict A useful XAI-based trust metric for arc fault classifiers, but the ground-truth definition is too loose to reproduce; worth reviewing with major revision. read the letter →

arxiv 2507.15239 v1 pith:KHLQ5UFP submitted 2025-07-21 cs.AI eess.SP

Qianchao Wang , Yuxuan Ding , Chuanzhen Jia , Zhe Li , Yaping Du This is my paper · ORCID
classification cs.AIeess.SP
keywords arcfaultdiagnosisexplainableartificialintelligencesoftevaluationindicatorSHAPocclusionsensitivityfeatureextractionscorelightweightneuralnetworkmodeltrustworthiness
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

The paper's central claim is that classification accuracy is not enough to certify an arc fault diagnosis model, because a model can reach high accuracy by exploiting dataset patterns unrelated to the fault. It proposes XSEI, a soft evaluation indicator that compares a model's explanation with a predefined ground truth of arc fault features: the five hand-picked features Variance, Entropy, Range, RMS, and Integral for feature-pool models, and the signal regions that change between normal and fault currents for deep models. The score is a Jaccard overlap between the model's attribution and this ground truth, so values near 1 mean the model's explanation coincides with the presumed physical fault. The paper argues these scores can guide model selection and ensemble weighting and reports experiments on two datasets with varying sample times and noise levels.

What carries the argument

The central object is a Jaccard-style soft score $g(f_i) \in [0,1]$ that measures overlap between an XAI attribution and the paper's ground-truth fault explanation. For feature-pool models the attribution comes from SHAP values, and for deep models it comes from occlusion sensitivity maps; the ground truth is either the five-feature set $S$ or the region-difference vector $r$. This score turns an explanation heatmap into one comparable number, which is what lets the paper rank models and propose the soft score as a basis for selection and ensemble weighting.

What would settle it

A decisive test would be to low-pass filter the arc-fault current so that the high-frequency components defining the ground-truth regions are removed; if a high-XSEI model still classifies faults accurately, the score does not measure reliance on the defined fault features.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes a definition of what a correct explanation of an arc fault should look like, then measures models against it. For feature-pool models the correct explanation is the set $S = \{\text{Variance}, \text{Entropy}, \text{Range}, \text{RMS}, \text{Integral}\}$; the model's explanation is the top-5 SHAP features, and the score is $g(f_i) = |S \cap S_{\mathrm{SHAP},i}| / |S \cup S_{\mathrm{SHAP},i}|$. For deep models the correct explanation is the regions $r$ where the arc fault current differs from the normal current, and the model's explanation is the regions $r_{\mathrm{Occlusion},i}$ found by occlusion sensitivity; the score is $g(f_i) = |r \cap r_{\mathrm{Occlusion},i}| / |r \cup r_{\mathrm{Occlusion},i}|$. Across the tested models, SEmodel and the L2/L1 norm score highest among machine learning methods, while AE and ArcNN score highest among deep learning methods, and several top-scoring models show a sharp accuracy drop when sample time or noise changes, which the authors read as evidence that the models are genuinely keyed to the fault features rather than to general patterns.

Load-bearing premise

The paper assumes its hand-picked feature set and region-difference rule are the true explanation of arc faults, and in particular that high-frequency current distortion is the causal mechanism; without a threshold for what counts as $x_n \neq \hat{x}_n$ in noisy data, the deep-learning ground truth is not operational on its own.

Editorial extensions

If this is right

  • If XSEI is right, a model with high accuracy but low soft score should be viewed with suspicion even when its test accuracy is excellent.
  • The soft score can be used as a basis for model selection when several models have comparable accuracy, as in the reported tables.
  • The scores can serve as ensemble weights, so models whose explanations match the ground truth contribute more to the combined decision.
  • A rapid drop in test accuracy as sample time lengthens or noise rises becomes a positive signal that the model relies on genuine fault features.
  • Average pooling is a better architectural choice than max pooling for locating arc faults, based on the soft score comparisons.

Reading between the lines

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

  • The region-difference ground truth in Equation (4) presumes the normal and fault signals can be aligned region by region; if real deployments have phase shifts or timing jitter, the score would penalize models that locate the same physical fault in shifted regions.
  • XSEI as defined measures agreement with the authors' chosen feature set, so a model could score high by using those features as statistical proxies rather than as causal fault indicators; an intervention that removes the high-frequency content while preserving the proxy feature values would separate the two cases.
  • The same overlap-scoring scheme could be transferred to other industrial diagnostic tasks where a small physical ground-truth feature set is known, but the definition of that ground truth would be the main obstacle.
  • Using XSEI as ensemble weights may over-reward models that match the chosen ground truth and under-reward models whose complementary errors would improve the ensemble; weighting should be validated against ensemble accuracy, not just soft scores.
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 / 7 minor

Summary. The paper proposes a soft evaluation indicator, XSEI, meant to assess whether AI-based arc fault diagnosis models actually rely on the true arc-fault features rather than on spurious patterns. For feature-pool models, XSEI is the Jaccard overlap between a hand-selected set of five features (Variance, Entropy, Range, RMS, Integral) and the top-5 SHAP features of the model. For deep learning models, XSEI is the Jaccard overlap between a ground-truth region mask defined by Eq. (4) and the region mask produced by occlusion sensitivity. The authors report test accuracy and XSEI scores for many ML and DL models over two experimental arc-fault datasets at several sample times and noise levels, and they propose a lightweight balanced neural network (LBNN) with average pooling. They conclude that high test accuracy does not guarantee trustworthiness, that a rapid drop in accuracy under lower precision is a sign that the model has found the key features, and that average pooling is preferable for arc fault diagnosis.

Significance. If the proposed evaluation indicator is valid, it would fill a genuine practical need: accuracy alone does not tell an engineer whether a fault detector is using diagnostic features or dataset artifacts. The paper deserves credit for using real arc-fault experiments following UL 1699 and GB 31143, for testing across many models and two datasets, and for presenting a simple, reproducible scoring formula. However, the central validity of XSEI rests entirely on the correctness of the authors' ground-truth definition of an arc-fault explanation. The deep-learning ground truth in Eq. (4) is not operational as stated (exact equality on noisy continuous signals), and the feature-pool ground-truth set is asserted rather than validated against any external standard. The evidence used to support XSEI is largely read from the same XSEI computations, making the validation circular. These issues are load-bearing for the paper's main claim.

major comments (4)
  1. [Section III-A2, Eq. (4)] The deep-learning ground truth r_n = 1 if x_n != x_hat_n is not operational for continuous noisy measurements. For two separately recorded current signals, exact equality has probability approximately zero under measurement noise, so r_n would be 1 for essentially every region; if the signals are aligned and quantized coarsely, r_n could be 0 everywhere. No threshold on the difference magnitude, no noise model, and no segmentation rule are supplied. Because Eq. (6) defines the XSEI score for all deep models, the scores reported in Tables II, III, IV, and V cannot be reproduced from the text and may be artifacts of an unspecified preprocessing decision.
  2. [Section III-A1] The ground-truth feature set S = {Variance, Entropy, Range, RMS, Integral} is selected by assertion rather than derived or externally validated. The text states that these features are 'normally' efficient for arc faults and then fixes them as ground truth; no evidence is given that exactly these five features, and not alternative candidates such as zero-crossing rate, spectral flatness, or wavelet coefficients, are the causal arc-fault features. Consequently, for feature-pool models, Eq. (5) measures agreement with a hand-picked set rather than with an independently established physical ground truth. A concrete external gold standard (expert-labeled fault segments, controlled fault injection, or a physics-based simulator) is needed to break this circularity.
  3. [Section IV-C and Conclusion (ii)] The heuristic that 'a rapid drop in test accuracy is a sign that the model has found the key features' is introduced as an interpretation of the same XSEI results and is then used as supporting evidence for XSEI and as the motivation for the average-pooling architecture in Section IV-E. No independent test, statistical significance, or out-of-sample prediction is provided; the fast-drop pattern is identified qualitatively in selected models (SEmodel, LightGBM, AE, ArcNN). This is circular validation: the phenomenon offered to validate the indicator is read off the indicator's own output. The paper should specify how the fast drop is measured and test it as a falsifiable prediction on held-out data.
  4. [Section III-B and Algorithm 1] The paper does not specify how the occlusion-derived region mask r_Occlusion,i is computed from the occlusion sensitivity heatmap. No threshold on the responsibility score in Eq. (2), no mask size or stride for M_j, and no rule for converting continuous attribution values into the binary mask used in Eq. (6) are given. Without these procedural details, the deep-model XSEI scores are not reproducible, since the same heatmap can yield many different binary masks under different thresholds.
minor comments (7)
  1. [Abstract] The abstract contains the typo 'the the correct explanation'.
  2. [Table I] The model listed as 'FW A-1DCNN' in Table I is referred to as 'IFW A-1DCNN' everywhere else; the table header should be corrected.
  3. [Section IV-C] The text 'SRN=-5' is a typo and should read 'SNR=-5'.
  4. [Section V] The conclusion says 'seven traditional machine learning methods and four deep learning methods', but Section IV lists nine ML models and five DL models; the counts are inconsistent.
  5. [Algorithm 1] In Algorithm 1, 'T op5' appears with an erroneous space, and the formatting of the g(f) vector and of the occlusion score accumulation is corrupted; these should be cleaned up.
  6. [Section II-B] The sentence 'when we mask all the signal, keeping nothing but Rmask' is confusingly worded and seems to contradict the condition in Eq. (3); the relationship between the responsibility score in Eq. (2) and the equivalence in Eq. (3) should be stated more carefully.
  7. [Figure 3] The text refers to Example 1 and Example 2, but the figure panels are labeled (a)-(d) without explicit cross-referencing; the reader must infer which panel corresponds to which example.

Circularity Check

0 steps flagged · score 2.0 of 10

XSEI is a definitional overlap metric against an explicitly stated ground truth; no circular reduction is exhibited, but the deep-learning ground truth lacks an operational threshold.

full rationale

The paper's central quantity, XSEI, is defined outright as a Jaccard overlap between each model's XAI attribution and the authors' explicitly chosen ground truth: Eq. 5 for feature-pool models and Eq. 6 for deep models. This is a definition, not a derived prediction, so there is no hidden reduction of a fitted quantity to a renamed output. The ground-truth feature set S = {Variance, Entropy, Range, RMS, Integral} is adopted from domain literature [15], [16], and the deep-learning ground truth r in Eq. 4 is the binary mask of regions where normal and arc-fault signals differ. The paper does not use XSEI to justify these ground truths; it uses XSEI to score models against them. The only self-citation found is reference [12] (Zhang et al., including co-authors Li, Jia, Du), cited for the KNN baseline; this is not load-bearing for the XSEI claim. The main weakness is not circularity but reproducibility: Eq. 4 uses exact equality on continuous noisy signals, and no threshold or binarization rule for occlusion heatmaps is provided, so the reported deep-model scores cannot be reproduced from the text. The paper's own stated limitation in Section V concerns the expertise required for classifier/XAI combinations, which is a practical concern, not a circularity admission. The 'fast drop' validation is an external behavioral check, not a re-statement of the XSEI formula. Overall, the derivation chain is self-contained and no circular step is exhibited; the deep-learning ground-truth operational gap is a correctness risk, which per the review rules is distinct from circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 1 invented entities

The central claim rests on definitions chosen in this paper rather than derived: the five-feature ground truth, the region-difference ground truth, the top-5 SHAP cutoff, and the unspecified occlusion mask configuration. The 'rapid accuracy drop' heuristic used to validate XSEI is an additional paper-specific assumption. No external gold standard or independent dataset is used, so these choices carry the weight of the conclusion.

free parameters (4)
  • Ground truth feature set S for feature-pool models = Variance, Entropy, Range, RMS, Integral
    Selected by the authors in Section III-A.1 without a formal selection rule; the XSEI score for ML models is measured as overlap with this set, so it is load-bearing.
  • Top-5 SHAP feature count = 5
    Section III-B uses top-5 SHAP features to compute S ∩ S_SHAP. No sensitivity analysis or justification is given for choosing exactly five.
  • Occlusion mask configuration = not specified
    Algorithm 1 and Section III-B use masks M and region scoring but never state mask width, stride, baseline (blur, constant, or noise), or the threshold for marking a region as a fault. The deep-learning scores depend on these choices.
  • LBNN architecture sizes = 6 and 16 filters, fully connected sizes 256 and 16
    Section IV-D Table VII: the convolutional filter counts and dense layer sizes are selected by hand to demonstrate the average-pooling idea, with no search or ablation beyond two modified models.
assumptions (5)
  • domain assumption The high-frequency content of the current signal is the causal basis for arc fault detection.
    Section III opening states this as the essential reason why ML models identify arc signals; it justifies the choice of ground truth features but is asserted, not derived.
  • ad hoc to paper The five selected statistical features (Variance, Entropy, Range, RMS, Integral) constitute the correct ground truth explanation for feature-pool models.
    Section III-A.1 chooses these features because they are 'normally' based on absolute or square values or signal bias; no independent evidence is provided that correct models must use exactly these.
  • ad hoc to paper The regions where normal and arc signals differ are the true arc fault features for deep learning models.
    Equation (4): r_n = 1 if x_n != x_hat_n. Since an arc fault changes the current waveform across most of the cycle, this definition conflates fault-related changes with load and noise differences.
  • ad hoc to paper A rapid drop in accuracy under lower sampling precision or higher noise indicates that a model is focusing on the true arc fault features.
    Section IV-C uses the fast-drop pattern as evidence of trustworthy feature extraction, but no formal argument or external validation is provided for this heuristic.
  • domain assumption SHAP and occlusion sensitivity correctly rank the features that a model actually uses for its decision.
    Section II relies on local feature attribution without discussing approximation error, baseline dependence, or whether SHAP and occlusion agree for the evaluated models.
invented entities (1)
  • Ground-truth correctness definition (five-feature set S and region-difference mask r)
    purpose: Serves as the reference standard against which XSEI measures every model, so it determines the meaning of 'correct explanation' for arc faults.
    The definition is asserted in Section III-A from a high-frequency causal story; no external measurement of actual fault location, physical verification, or independent expert benchmark is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explainable Artificial Intelligence based Soft Evaluation Indicator for Arc Fault Diagnosis." pith.science (2026). https://pith.science/paper/KHLQ5UFP

@misc{pith2026250715239,
  author       = {Pith},
  title        = {Pith review of: Explainable Artificial Intelligence based Soft Evaluation Indicator for Arc Fault Diagnosis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KHLQ5UFP}},
  note         = {Machine review of arXiv:2507.15239}
}
read the original abstract

Novel AI-based arc fault diagnosis models have demonstrated outstanding performance in terms of classification accuracy. However, an inherent problem is whether these models can actually be trusted to find arc faults. In this light, this work proposes a soft evaluation indicator that explains the outputs of arc fault diagnosis models, by defining the the correct explanation of arc faults and leveraging Explainable Artificial Intelligence and real arc fault experiments. Meanwhile, a lightweight balanced neural network is proposed to guarantee competitive accuracy and soft feature extraction score. In our experiments, several traditional machine learning methods and deep learning methods across two arc fault datasets with different sample times and noise levels are utilized to test the effectiveness of the soft evaluation indicator. Through this approach, the arc fault diagnosis models are easy to understand and trust, allowing practitioners to make informed and trustworthy decisions.

Figures

Figures reproduced from arXiv: 2507.15239 by the authors.

Figure 1
Figure 1. The schematic diagram of the low-voltage soft Evaluation Process. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Experimental platform of arc fault experiment. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Data analysis of experimental dataset. from arc fault currents with a single load [12]. Whereas, the core obstacle in arc fault diagnosis is the same, that is, finding the high-frequency fault signal in the current signal without being affected by noise and sampling rate. In addition, since the laboratory experiments are relatively simple, to simulate more practical and complex engineering problems, we add additiona… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Visualization examples of soft feature extraction scores using datasets [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 33 canonical work pages

  1. [1]

    Machine learning approach to detect arc faults based on regular coupling features,

    R. Jiang, G. Bao, Q. Hong, and C. D. Booth, “Machine learning approach to detect arc faults based on regular coupling features,” IEEE Transactions on Industrial Informatics , vol. 19, no. 3, pp. 2761–2771, 2022

    work page 2022

  2. [2]

    Advancements in Arc Fault Detection for Electrical Distribution Systems: A Comprehensive Review from Artificial Intelligence Perspective

    K. Thakur, D. Dwivedi, K. Babu, A. M. Parimi, P. K. Yemula, P. Chakraborty, and M. Pal, “Advancements in arc fault detection for electrical distribution systems: A comprehensive review from artificial intelligence perspective,” arXiv preprint arXiv:2311.16804 , 2023

    work page Pith review arXiv 2023

  3. [3]

    A coupling method for identifying arc faults based on short-observation-window svdr,

    R. Jiang, G. Bao, Q. Hong, and C. D. Booth, “A coupling method for identifying arc faults based on short-observation-window svdr,” IEEE Transactions on Instrumentation and Measurement , vol. 70, pp. 1–10, 2021. 10

    work page 2021

  4. [4]

    Series dc arc fault detection based on ensemble machine learning,

    V . Le, X. Yao, C. Miller, and B.-H. Tsao, “Series dc arc fault detection based on ensemble machine learning,” IEEE Transactions on Power Electronics, vol. 35, no. 8, pp. 7826–7839, 2020

    work page 2020

  5. [5]

    Design of a dc series arc fault detector for photovoltaic system protection,

    J.-C. Gu, D.-S. Lai, J.-M. Wang, J.-J. Huang, and M.-T. Yang, “Design of a dc series arc fault detector for photovoltaic system protection,”IEEE Transactions on Industry Applications , vol. 55, no. 3, pp. 2464–2471, 2019

    work page 2019

  6. [6]

    Real-time dc series arc fault detection based on noise pattern analysis in photovoltaic system,

    J.-B. Ahn, H.-B. Jo, and H.-J. Ryoo, “Real-time dc series arc fault detection based on noise pattern analysis in photovoltaic system,” IEEE Transactions on Industrial Electronics , vol. 70, no. 10, pp. 10 680– 10 689, 2023

    work page 2023

  7. [7]

    An integrated dc series arc fault detection method for different operating conditions,

    Z. Yin, L. Wang, B. Zhang, L. Meng, and Y . Zhang, “An integrated dc series arc fault detection method for different operating conditions,” IEEE Transactions on Industrial Electronics, vol. 68, no. 12, pp. 12 720– 12 729, 2021

    work page 2021

  8. [8]

    An enhanced cassie–mayr-based approach for dc series arc modeling in pv systems,

    M. Jalil, H. Samet, T. Ghanbari, and M. Tajdinian, “An enhanced cassie–mayr-based approach for dc series arc modeling in pv systems,” IEEE Transactions on Instrumentation and Measurement , vol. 70, pp. 1–10, 2021

    work page 2021

Show all 36 references
  1. [9]

    Series arc fault detection using regular sig- nals and time-series reconstruction,

    R. Jiang and Y . Zheng, “Series arc fault detection using regular sig- nals and time-series reconstruction,” IEEE Transactions on Industrial Electronics, vol. 70, no. 2, pp. 2026–2036, 2023

    2026

  2. [10]

    Feature extraction and power quality disturbances classification using smart meters signals,

    F. A. S. Borges, R. A. S. Fernandes, I. N. Silva, and C. B. S. Silva, “Feature extraction and power quality disturbances classification using smart meters signals,” IEEE Transactions on Industrial Informatics , vol. 12, no. 2, pp. 824–833, 2016

    2016

  3. [11]

    Series arc detection and complex load recognition based on principal component analysis and support vector machine,

    J. Jiang, Z. Wen, M. Zhao, Y . Bie, C. Li, M. Tan, and C. Zhang, “Series arc detection and complex load recognition based on principal component analysis and support vector machine,” IEEE Access, vol. 7, pp. 47 221–47 229, 2019

    2019

  4. [12]

    Lightweight ac arc fault detection method by integration of event-based load classification,

    Y . Zhang, H. C. Chen, Z. Li, C. Jia, Y . Du, K. Zhang, and H. Wei, “Lightweight ac arc fault detection method by integration of event-based load classification,” IEEE Transactions on Industrial Electronics, vol. 71, no. 4, pp. 4130–4140, 2024

    2024

  5. [13]

    Ac series arc fault detection based on rlc arc model and convolutional neural network,

    R. Jiang, Y . Wang, X. Gao, G. Bao, Q. Hong, and C. D. Booth, “Ac series arc fault detection based on rlc arc model and convolutional neural network,” IEEE Sensors Journal , vol. 23, no. 13, pp. 14 618–14 627, 2023

    2023

  6. [14]

    Deep learning-based series ac arc detection algorithms,

    C.-J. Park, H.-L. Dang, S. Kwak, and S. duck Choi, “Deep learning-based series ac arc detection algorithms,” Journal of Power Electronics, vol. 21, pp. 1621 – 1631, 2021. [Online]. Available: https://api.semanticscholar.org/CorpusID:238670401

    2021

  7. [15]

    Series ac arc fault detection method based on hybrid time and frequency analysis and fully connected neural network,

    Y . Wang, F. Zhang, X. Zhang, and S. Zhang, “Series ac arc fault detection method based on hybrid time and frequency analysis and fully connected neural network,”IEEE Transactions on Industrial Informatics, vol. 15, no. 12, pp. 6210–6219, 2019

    2019

  8. [16]

    Intever public database for arcing event detection: Feature analysis, benchmark test, and multi-scale cnn application,

    G. Zhou, L. Huang, Z. Li, H. Tian, B. Zhang, M. Fu, Y . Feng, and C. Huang, “Intever public database for arcing event detection: Feature analysis, benchmark test, and multi-scale cnn application,” IEEE Transactions on Instrumentation and Measurement , vol. 70, pp. 1–15, 2021

    2021

  9. [17]

    An on-line detection method and device of series arc fault based on lightweight cnn,

    Z. Wang, S. Tian, H. Gao, C. Han, and F. Guo, “An on-line detection method and device of series arc fault based on lightweight cnn,” IEEE Transactions on Industrial Informatics, vol. 19, no. 10, pp. 9991–10 003, 2023

    2023

  10. [18]

    Arcnet: Series ac arc fault detection based on raw current and convolutional neural network,

    Y . Wang, L. Hou, K. C. Paul, Y . Ban, C. Chen, and T. Zhao, “Arcnet: Series ac arc fault detection based on raw current and convolutional neural network,” IEEE Transactions on Industrial Informatics , vol. 18, no. 1, pp. 77–86, 2022

    2022

  11. [19]

    A simplified current feature extraction and deployment method for dc series arc fault detection,

    J. Yan, Q. Li, and S. Duan, “A simplified current feature extraction and deployment method for dc series arc fault detection,” IEEE Transactions on Industrial Electronics , vol. 71, no. 1, pp. 625–634, 2024

    2024

  12. [20]

    Causal inference,

    M. A. Hern ´an and J. M. Robins, “Causal inference,” 2010

    2010

  13. [21]

    Explainable artificial intelligence (xai): What we know and what is left to attain trustworthy artificial intelligence,

    S. Ali, T. Abuhmed, S. El-Sappagh, K. Muhammad, J. M. Alonso- Moral, R. Confalonieri, R. Guidotti, J. Del Ser, N. D ´ıaz-Rodr´ıguez, and F. Herrera, “Explainable artificial intelligence (xai): What we know and what is left to attain trustworthy artificial intelligence,” Inform...

    2023

  14. [22]

    Harnessing explainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of grad-cam and shap,

    C. Van Zyl, X. Ye, and R. Naidoo, “Harnessing explainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of grad-cam and shap,” Applied Energy, vol. 353, p. 122079, 2024

    2024

  15. [23]

    Interpretable machine learning for building energy management: A state-of-the-art review,

    Z. Chen, F. Xiao, F. Guo, and J. Yan, “Interpretable machine learning for building energy management: A state-of-the-art review,” Advances in Applied Energy , vol. 9, p. 100123, 2023

    2023

  16. [24]

    Toward transparent load disaggregation—a framework for quantitative evaluation of explainabil- ity using explainable ai,

    D. Batic, V . Stankovic, and L. Stankovic, “Toward transparent load disaggregation—a framework for quantitative evaluation of explainabil- ity using explainable ai,” IEEE Transactions on Consumer Electronics , vol. 70, no. 1, pp. 4345–4356, 2024

    2024

  17. [25]

    Aiming to complex power quality disturbances: A novel decomposition and detection framework,

    K. Zhu, Z. Teng, W. Qiu, A. Mingotti, Q. Tang, and W. Yao, “Aiming to complex power quality disturbances: A novel decomposition and detection framework,” IEEE Transactions on Industrial Informatics , vol. 20, no. 3, pp. 4317–4326, 2024

    2024

  18. [26]

    A unified approach to interpreting model predictions,

    S. M. Lundberg and S.-I. Lee, “A unified approach to interpreting model predictions,” Advances in neural information processing systems, vol. 30, 2017

    2017

  19. [27]

    Algorithms to es- timate shapley value feature attributions,

    H. Chen, I. C. Covert, S. M. Lundberg, and S.-I. Lee, “Algorithms to es- timate shapley value feature attributions,” Nature Machine Intelligence , vol. 5, no. 6, pp. 590–601, 2023

    2023

  20. [28]

    Visualizing and understanding convo- lutional networks,

    M. D. Zeiler and R. Fergus, “Visualizing and understanding convo- lutional networks,” in Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part I 13. Springer, 2014, pp. 818–833

    2014

  21. [29]

    The many shapley values for model explanation,

    M. Sundararajan and A. Najmi, “The many shapley values for model explanation,” in International conference on machine learning. PMLR, 2020, pp. 9269–9278

    2020

  22. [30]

    Occlusion sensitivity analysis with augmentation subspace perturbation in deep feature space,

    P. H. Valois, K. Niinuma, and K. Fukui, “Occlusion sensitivity analysis with augmentation subspace perturbation in deep feature space,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024, pp. 4829–4838

    2024

  23. [31]

    Series ac arc fault detection method based on l2/l1 norm and classification algorithm,

    W. Dai, X. Zhou, Z. Sun, Q. Miao, and G. Zhai, “Series ac arc fault detection method based on l2/l1 norm and classification algorithm,” IEEE Sensors Journal , vol. 24, no. 10, pp. 16 661–16 672, 2024

    2024

  24. [32]

    Lightgbm: A highly efficient gradient boosting decision tree,

    G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T.- Y . Liu, “Lightgbm: A highly efficient gradient boosting decision tree,” Advances in neural information processing systems , vol. 30, 2017

    2017

  25. [33]

    Xgboost: A scalable tree boosting system,

    T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining , 2016, pp. 785–794

    2016

  26. [34]

    Series arc fault identification based on a novel newton-optimized sparse kernel support vector machine,

    H. Hua, D. Fang, and M. Guo, “Series arc fault identification based on a novel newton-optimized sparse kernel support vector machine,” IEEE Transactions on Instrumentation and Measurement , vol. 74, pp. 1–13, 2025

    2025

  27. [35]

    Research on the identification method of series arc fault based on the feature sensitivity analysis and the tva coefficient optimized random forest,

    H. Tong, X. Zeng, K. Yu, J. Mu, C. Luo, and B. Liu, “Research on the identification method of series arc fault based on the feature sensitivity analysis and the tva coefficient optimized random forest,” IEEE Transactions on Power Delivery , vol. 39, no. 2, pp. 751–762, 2024

    2024

  28. [36]

    A novel series arc fault detection method based on ceemdan and ifaw-1dcnn,

    J. Wu, W. Wang, T. Shang, and J. Cao, “A novel series arc fault detection method based on ceemdan and ifaw-1dcnn,” IEEE Transactions on Dielectrics and Electrical Insulation , vol. 31, no. 2, pp. 1020–1029, 2024

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.