Pith. sign in

REVIEW 5 major objections 4 minor 37 references

Interpretable Recognition of Fused Magnesium Furnace Working Conditions with Deep Convolutional Stochastic Configuration Networks

T0 review · 5 major / 4 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 A stochastic-configuration deep network recognizes fused magnesium furnace working conditions from images with 92.57% test accuracy and heatmap-based explanations, without backpropagation.

desk verdict A plausible industrial monitoring paper whose new RL-pruning module is interesting, but the convergence proof is broken and key reported numbers are inconsistent. read the letter →

arxiv 2501.02740 v1 pith:AKXDN4W2 submitted 2025-01-06 cs.CV cs.AI

Li Weitao , Zhang Xinru , Wang Dianhui , Tong Qianqian , Chai Tianyou This is my paper · ORCID
classification cs.CVcs.AI
keywords fusedmagnesiumfurnaceworkingconditionrecognitiondeepconvolutionalstochasticconfigurationnetworksGaussiandifferentialconvolutionkernelclassactivationmappingreinforcementlearningpruninginterpretabilityindustrialimageclassification
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 proposes a way to recognize the working condition of a fused magnesium furnace—normal, underburn, overheating, or abnormal exhaust—from furnace-mouth images, and claims it is both more accurate and more explainable than existing deep-learning approaches. The central idea is to build a deep convolutional network without backpropagation: convolutional kernels are generated one by one as Gaussian-difference filters, and a supervised rule accepts only kernels that reduce the current recognition error. The authors report that the resulting model reaches 92.57% test accuracy, that its class activation maps highlight the correct region of the furnace, and that a reinforcement-learning step prunes redundant kernels to make the model compact. If this holds, furnace operators could trust an automatically generated visual explanation rather than relying only on manual inspection.

What carries the argument

The load-bearing object is the Gaussian differential convolution kernel, defined as $$\psi(x,y)=\frac{1}{2\pi}\left($e^{{-\frac{x^2+y^2}}${2\$xi^{2}$}}-\frac{1}{r}$e^{{-\frac{x^2+y^2}}${$2r^{2}$\$xi^{2}$}}\right),$$ a center-surround filter that sharpens edges and texture at the furnace mouth; $\xi$ controls filter width and $r$ controls edge scale. The construction mechanism is a supervised configuration loop: candidate kernels are drawn uniformly over $\xi\in[0.5,5]$ and $r\in[0.8,1.5]$, and a candidate is kept only if its convergence score is positive, which is meant to guarantee a contraction of the residual error. The network grows incrementally—one kernel at a time, then one layer at a time—with output weights fitted by least squares, avoiding backpropagation. Interpretability comes from the channel feature independence coefficient $FC^l_\varrho$, which weights feature maps in a class activation map $L_q$; trustworthiness is scored by the IoU of the highlighted region with expert annotation. Finally, a DDPG reinforcement learner chooses per-layer pruning ratios under a joint reward $R = \mathrm{ACC}_{\mathrm{val}} + \mathrm{IoU}_{\mathrm{val}} - \beta P_A$.

What would settle it

Record the residual norm after each accepted kernel during the incremental DCSCN construction on the reported 12,000-image dataset; if any accepted kernel violates the paper's contraction inequality or the per-layer error does not decrease monotonically under the stated parameter ranges, the claimed convergence guarantee is refuted. A behavioral counterpart: test the pruned 8-layer model on a fresh labeled set of furnace videos from another plant and measure both accuracy and class activation map IoU against expert annotations.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that stochastic configuration networks—randomized learners whose hidden parameters are selected by a supervisory criterion—can be extended to deep convolutional networks for industrial image recognition. The paper argues that Gaussian differential convolution kernels, generated randomly within a data-correlated range and accepted only when they satisfy a residual-based inequality, let a multi-layer convolutional network converge in error layer by layer without gradient descent. The same construction yields feature maps whose channel independence scores can be combined with class scores into class activation maps, giving an interpretable heatmap of where the network looks. The paper further claims that pruning kernels with a DDPG reinforcement learner, guided by a reward combining accuracy, interpretability trustworthiness (IoU against expert annotations), and parameter count, improves focus and keeps accuracy high. Tested on 12,000 augmented images of four furnace conditions, the method's 92.57% test accuracy is claimed to beat CNNs, Bayesian networks, Swin Transformer, CNN+LSTM, SCNs, 2DSCNs, and DeepSCNs.

Load-bearing premise

The convergence guarantee assumes that every newly accepted Gaussian-difference kernel shrinks the training error by a fixed factor; with the paper's own ranges $\xi\in[0.5,5]$ and $r\in[0.8,1.5]$, that factor is not always forced below one, so the error-decrease premise can fail.

Editorial extensions

If this is right

  • If the central claim is right, furnace operators can receive both a class label and a heatmap showing which part of the furnace mouth drove the decision, making each alarm checkable by a human.
  • Because convolutional kernels are configured incrementally rather than trained by backpropagation, the method avoids learning-rate sensitivity and weight-initialization dependence that complicate conventional CNN training on small industrial datasets.
  • The RL pruning step selects kernel subsets that preserve accuracy and interpretability while cutting parameters and training time, so the deployed model is smaller and faster than the unpruned network.
  • The same supervised kernel-configuration strategy can build deep convolutional recognizers for other process industries where labeled abnormal samples are scarce and explainable decisions are required.

Reading between the lines

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

  • Our reading: the convergence proof is the part to test first. With $\xi\in[0.5,5]$ and $r\in[0.8,1.5]$, $\xi+r$ is often larger than 1, so the contraction factor in the paper's inequality need not be below 1; the claimed monotone decrease of the residual thus depends on a premise the stated parameter ranges do not guarantee.
  • The paper reports single-run recognition results; a repeated-seed or cross-factory evaluation would tell whether the 2–4 point accuracy gap over CNN, Swin Transformer, and CNN+LSTM is stable rather than an artifact of one split.
  • The interpretability claim could be stress-tested by having operators identify the anomaly region from the class activation map alone; the IoU metric already approximates this, but a human study would connect the quantitative score to real usability.
  • Because the Gaussian-difference kernel family is fixed in advance, the method's "physical meaning" is inherited from the filter design; one could ask whether allowing the kernel family to be learned while keeping the stochastic configuration rule would retain interpretability and improve accuracy on larger datasets.
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

5 major / 4 minor

Summary. The paper proposes deep convolutional stochastic configuration networks (DCSCNs) for recognizing working conditions of fused magnesium furnaces from furnace images. The method generates Gaussian differential convolution kernels through a supervised stochastic configuration mechanism, constructs a deep convolutional network incrementally without backpropagation, claims a global convergence proof for the resulting errors, and defines a class-activation-map-based interpretability index (IoU) together with a DDPG-based reinforcement-learning pruning procedure driven by a joint reward of accuracy, interpretability, and parameter count. Experiments on a 12,000-image industrial dataset report a test accuracy of 92.57%, outperforming several comparison methods, with qualitative CAM visualizations claimed to locate the relevant furnace regions. The central theoretical claim of hierarchical global convergence, however, is not established as written: the contraction argument in Section III.B.3 contains an invalid inequality and an algebraic error, and the interpretability metric defined in Section III.C is never reported quantitatively.

Significance. If the convergence proof were correct and the interpretability index were properly evaluated, the paper would offer a useful contribution to industrial AI by combining a backpropagation-free randomized construction of deep CNNs with RL-based pruning and visual explainability. The empirical accuracy improvement on a real fused-magnesium-furnace dataset is credible and potentially valuable to practitioners. However, the paper's stated theoretical contribution is load-bearing and is not merely incomplete; the contraction inequality used to prove monotone convergence is numerically false under the reported parameter ranges, and the projection step in Eq. (22) is algebraically wrong. In addition, the interpretability evaluation is qualitative only, and the reported parameter counts are inconsistent across tables. The strengths are the clear application framing, the use of physically meaningful Gaussian-differential kernels, and the reproducible experimental setup; the weaknesses are in the formal guarantees and the quantitative validation of interpretability, which are central to the paper's claims.

major comments (5)
  1. [Section III.B.3, Eq. (19)] The asserted inequality ||e^l_{C_l}||^2 ≤ (ξ+r) u_{C_l} ||e^l_{C_l-1}||^2 is not a contraction under the stated parameter ranges. With ξ∈[0.5,5], r∈[0.8,1.5], and u_C=1/C, the factor (ξ+r)u_C is at least 1.3 for C=1 and can be as large as 6.5, so the right-hand side can exceed the left-hand side even when the error has not improved. The chain in Eq. (23) therefore does not imply monotone decrease of the recognition error. Since Contribution 1 explicitly claims a proof of global convergence, the paper cannot discharge this step by citing [25]; it must either prove Eq. (19) under conditions that make the factor strictly less than 1, or restrict the parameter ranges accordingly.
  2. [Section III.B.3, Eq. (22)] The equality ||e^l_{C_l} − \bar{O}^{l+1}_1 A^{l+1}_1||^2 = ||e^l_{C_l}||^2 − Σ_{q=1}^m \bar{O}^{l+1}_{1,q} is algebraically wrong. The correct orthogonal projection identity gives ||e − \bar{O} A||^2 = ||e||^2 − Σ_q ⟨e_q, A_q⟩^2 / ||A_q||^2, not subtraction of the output weight \bar{O} itself. The written equality is dimensionally inconsistent, and the inequality ≤ ||e^l_{C_l}||^2 does not follow from it. This step is the bridge from layer l to layer l+1, so the claimed monotone convergence is not established.
  3. [Section III.C and Section IV.C.2] The IoU-based interpretable trustworthiness index is defined in Eqs. (28)-(29) and used as a component of the RL reward in Eq. (30), but no numerical IoU values are reported anywhere in the experimental section. The interpretability claim (“accurately locate the target region of interest”) is supported only by qualitative CAM images. In addition, computing IoU requires a procedure for extracting the highlighted region d_i from the CAM (e.g., a thresholding rule), which is not specified; without that procedure the metric is not well-defined. Because IoU_val is also fed into the reward, a separate held-out evaluation of IoU would be needed to avoid circularity; the paper does not provide one.
  4. [Section IV.D, Tables I-III] The “Proposed method” row in Table I reports P_A = 48.85 MB, while Tables II and III report P_A = 12.854 MB for the same method and the same test accuracy 92.57%. Section IV.C.1 attributes 94.53%/92.78% to the 8-layer model, which matches the “Without adaptive kernel pruning mechanisms” row of Table I rather than the “Proposed method” row. These discrepancies are material to the model-compression claim; the authors should state which configuration (before or after RL pruning, and with which kernel assignment) is used in each table, and ensure the numbers are consistent.
  5. [Section III.B.1 and III.B.2] The kernel selection conditions are internally inconsistent. Eq. (8) uses p u_C b^2 ||e||^2, Eq. (9) uses ξ u_C b^2 ||e||^2, but the convergence score in Eq. (17) uses (ξ+r)u_C ||e||^2 with no b^2 and with a different constant structure. The paper never defines how p, b, ξ, r, and u_C jointly guarantee the inequality in Eq. (19). Without a precise link between the selection criterion and the asserted contraction inequality, the hierarchical convergence claim is not supported.
minor comments (4)
  1. [Section IV.C.2 and Section V] The caption of Fig. 11 contains the typo “convolutinal”, and the conclusion contains “practical species for fused magnesium furnace working condition recognition”, which should read “practical support for ...”. These should be corrected.
  2. [Section IV.C.3 and Eqs. (31)-(35)] The symbol γ is used both for the DDPG discount factor in Eq. (32) and for the learning rate in the sentence “the learning rate γ is 0.005”, while Eqs. (34)-(35) use τ for the soft-update rate. Distinct symbols should be used for these three quantities.
  3. [Section III.B] The notation C is overloaded: C indexes convolutional kernels (C=1,...,C_l), C_l denotes the number of kernels in layer l, and C_{l-1} denotes the number of input channels. In Eq. (26) the feature map dimensions are written as H×W×C_l, while the text also uses C_l for the number of channels. Please disambiguate the two uses.
  4. [Section III.B.1] The proof claims lim_{l→+∞} ||F − F^l|| = 0, but the construction stops at finite L_max and C_l,max. The asymptotic statement should be replaced by a finite-sample error bound consistent with the stopping criterion, or the precise sense of the convergence claim should be stated.

Circularity Check

1 steps flagged · score 4.0 of 10

The convergence proof rests on Eq. (19), which is imported as 'can be derived ... [25]' from the authors' own earlier DCSCN paper; the 92.57% empirical accuracy is independent.

  1. self citation load bearing [Section III.B.3, Eq. (19)]
    "Assuming that the number of convolutional kernels in the lth layer of DCSCNs is Cl, it can be derived that the lth layer convolution satisfies [25]: ||e_l_Cl||^2 ≤ (ξ+r)u_Cl ||e_l_Cl−1||^2 (19)"

    Equation (19) is the only inequality that powers the monotone-decrease chain in Eq. (23), and that chain is what the paper presents as its 'rigorous theoretical proof of convergence' and as contribution 1. The bound is not proved in this manuscript; it is simply asserted as 'it can be derived ... [25]', where [25] (W. Li, Q. Tong, D. Wang, G. Wu) is the authors' own prior DCSCN paper. Thus the paper's central theoretical guarantee is inherited from a self-citation rather than established by the present derivation. Even taken as given, with ξ ∈ [0.5, 5], r ∈ [0.8, 1.5], and u_1 = 1, the factor (ξ+r)u_C need not be less than 1, so the cited inequality does not itself force contraction; moreover, the follow-up projection step in Eq. (22) is algebraically incorrect.

full rationale

The empirical claims are not circular: the 92.57% test accuracy is measured on a held-out 20% test set (Section IV.A, Table II), and the compared baselines are external methods trained and tested separately. The RL pruning and CAM evaluations are likewise not fitted to the test labels used for the headline result. The IoU-based reward in Eq. (30) is a mild self-reference, but the paper's published interpretability evidence is largely qualitative CAM inspection and no final IoU score is reported as the headline evaluation metric, so this does not by itself constitute a circular prediction. The only load-bearing circularity is in the theory: the convergence proof of DCSCNs, listed as contribution 1, depends entirely on Eq. (19), which is not derived here but is imported from the authors' own prior paper [25]. Since that inequality is precisely the contraction step yielding Eq. (23) and the claimed monotone error decrease, the 'proof' reduces at its pivotal point to a self-citation. Separately, Eq. (22)'s projection identity is misstated and the stated parameter ranges do not force (ξ+r)u_C < 1; those correctness problems reinforce that the convergence guarantee is not established as written, though they are not themselves circularity. Score 4: some self-citation is load-bearing, but the empirical contribution remains independently testable.

Assumptions & free parameters 9 free parameters · 6 assumptions · 0 invented entities

The central claim rests mainly on the SCN universal approximation property (imported from prior work), on the Gaussian differential kernel family, and on the contraction inequality assumed from [25]. The RL reward incorporates the interpretability index, making the interpretability evaluation not independent of the training objective. No new physical entities are introduced; the free parameters are mostly hand-chosen hyperparameters whose values are needed to reproduce the method.

free parameters (9)
  • Gaussian kernel scale xi sampling range = [0.5, 5]
    Uniform random kernel width range chosen by hand (Section IV.C.1); controls texture scale and appears in the convergence condition.
  • Gaussian kernel scale factor r sampling range = [0.8, 1.5]
    Hand-chosen range for the difference-of-Gaussians scale ratio (Section IV.C.1).
  • Desired error limit e_bar = 0.01
    Stopping threshold for incremental kernel generation (Section IV.C.1).
  • Candidate kernel count Tmax = 100
    Number of random candidate kernels tested per added node (Section IV.C.1).
  • Max layers Lmax and max kernels per layer = 10 / 50
    Network architecture caps chosen by hand (Section IV.C.1).
  • Contraction sequence u_C = 1/C
    Chosen sequence; the convergence proof depends on it (Section III.B.1 and IV.C.1).
  • RL reward weight beta = not reported
    Weight in joint reward Eq (30); absence prevents reproduction of the pruning behavior.
  • DDPG learning rate and discount = 0.005 / 0.9
    RL hyperparameters from Section IV.C.3; note gamma is used for both discount and learning rate, a notation clash.
  • SCN constants p and b = not specified
    Appear in kernel selection inequality Eq (8), but their values are never given; the theorem also requires 0 < ||A|| < b.
assumptions (6)
  • standard math The function space span(Gamma) is dense in L2 and bounded kernels satisfy 0 < ||A|| < b
    Invoked before Eq (8) in Section III.B.1 to license residual-based kernel selection; proven for SCNs in prior work [29].
  • domain assumption Real-world furnace images are separable by Gaussian differential kernels
    Section III.B.2 asserts this because furnace regions have brightness discontinuities; no quantitative support is provided.
  • domain assumption Convolutional kernels in good deep networks should follow a Gaussian distribution
    Section III.B.2 cites [31] as the basis for choosing the Gaussian differential kernel family.
  • ad hoc to paper The inequality of Eq (19) is inherited from reference [25]
    The convergence proof starts from this cited inequality instead of proving it for the constructed kernels.
  • domain assumption Random train/test split treats video frames as independent samples
    Section IV.A describes a 60/20/20 random split of frame-separated images; adjacent frames from the same video are likely correlated, so independence may fail.
  • domain assumption IoU between a heatmap highlight region and expert annotation measures interpretability trustworthiness
    Section III.C defines IoU for this purpose, but the threshold for the highlight region is unspecified and the values are never reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Interpretable Recognition of Fused Magnesium Furnace Working Conditions with Deep Convolutional Stochastic Configuration Networks." pith.science (2026). https://pith.science/paper/AKXDN4W2

@misc{pith2026250102740,
  author       = {Pith},
  title        = {Pith review of: Interpretable Recognition of Fused Magnesium Furnace Working Conditions with Deep Convolutional Stochastic Configuration Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AKXDN4W2}},
  note         = {Machine review of arXiv:2501.02740}
}
read the original abstract

To address the issues of a weak generalization capability and interpretability in working condition recognition model of a fused magnesium furnace, this paper proposes an interpretable working condition recognition method based on deep convolutional stochastic configuration networks (DCSCNs). Firstly, a supervised learning mechanism is employed to generate physically meaningful Gaussian differential convolution kernels. An incremental method is utilized to construct a DCSCNs model, ensuring the convergence of recognition errors in a hierarchical manner and avoiding the iterative optimization process of convolutional kernel parameters using the widely used backpropagation algorithm. The independent coefficient of channel feature maps is defined to obtain the visualization results of feature class activation maps for the fused magnesium furnace. A joint reward function is constructed based on the recognition accuracy, the interpretable trustworthiness evaluation metrics, and the model parameter quantity. Reinforcement learning (RL) is applied to adaptively prune the convolutional kernels of the DCSCNs model, aiming to build a compact, highly performed and interpretable network. The experimental results demonstrate that the proposed method outperforms the other deep learning approaches in terms of recognition accuracy and interpretability.

Figures

Figures reproduced from arXiv: 2501.02740 by the authors.

Figure 1
Figure 1. Structure of interpretable fused magnesium furnace working condition recognition model based on deep convolutional [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Deep convolutional stochastic configuration network structure diagram. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Schematic diagram of the class activation mapping [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (11 more)
Figure 4
Figure 4. Figure 4: Structure of convolutional kernel adaptive pruning [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Results after image data enhancement for normal working conditions. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Results after image data enhancement for underburn working conditions. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Results after image data enhancement for overheating working conditions. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Results after image data enhancement for abnormal exhaust working conditions. [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 10
Figure 10. Figure 10: Recognition accuracy curve of multi-layer DCSCNs [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]
Figure 9
Figure 9. Figure 9: Single-layer DCSCNs recognition accuracy curves with [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 12
Figure 12. Figure 12: Class activation mapping for different working conditions. [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]
Figure 13
Figure 13. Figure 13: The average reward curve of the RL training process. [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]
Figure 15
Figure 15. Figure 15: Recognition accuracy curves of training samples for [PITH_FULL_IMAGE:figures/full_fig_p013_15.png]
Figure 14
Figure 14. Figure 14: Comparison of the algorithm proposed in this paper [PITH_FULL_IMAGE:figures/full_fig_p013_14.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 35 canonical work pages

  1. [25]

    Research on working condition recognition method of fmf based on deep convolutional stochastic configuration network,

    W. Li, Q. Tong, D. Wang, and G. Wu, “Research on working condition recognition method of fmf based on deep convolutional stochastic configuration network,” Acta Automatica Sinica, vol. 50, no. 3, pp. 527– 544, 2024

    work page 2024

  2. [1]

    Disturbance robust abnormality diagnosis of fused magnesium furnaces using deep neural networks,

    Q. Liu, Y . Zhang, G. Wu, and Z. Fan, “Disturbance robust abnormality diagnosis of fused magnesium furnaces using deep neural networks,” IEEE Transactions on Artificial Intelligence , vol. 4, no. 4, pp. 669–678, 2023

    work page 2023

  3. [2]

    Demand forecasting of the fused magnesia smelting process with system identification and deep learning,

    T. Chai, J. Zhang, and T. Yang, “Demand forecasting of the fused magnesia smelting process with system identification and deep learning,” IEEE Transactions on Industrial Informatics , vol. 17, no. 12, pp. 8387– 8396, 2021

    work page 2021

  4. [3]

    Intelligent demand forecasting of smelting process using data-driven and mechanism model,

    J. Yang, T. Chai, C. Luo, and W. Yu, “Intelligent demand forecasting of smelting process using data-driven and mechanism model,” IEEE Transactions on Industrial Electronics , vol. 66, no. 12, pp. 9745–9755, 2019

    work page 2019

  5. [4]

    Data-driven abnormal condition identification and self-healing control system for fused magnesium furnace,

    Z. Wu, Y . Wu, T. Chai, and J. Sun, “Data-driven abnormal condition identification and self-healing control system for fused magnesium furnace,” IEEE Transactions on Industrial Electronics , vol. 62, no. 3, pp. 1703–1715, 2015

    work page 2015

  6. [5]

    Signal-compensation-based adaptive PID control for fused magnesia smelting processes,

    W. Wang, T. Chai, H. Wang, and Z. Wu, “Signal-compensation-based adaptive PID control for fused magnesia smelting processes,” IEEE Transactions on Industrial Electronics , vol. 70, no. 9, pp. 9441–9451, 2023

    work page 2023

  7. [6]

    Abnormal condition identification and safe control scheme for the electro-fused magnesia smelting process,

    H. Li, F. Wang, and H. Li, “Abnormal condition identification and safe control scheme for the electro-fused magnesia smelting process,” ISA Transactions, vol. 76, pp. 178–187, 2018

    work page 2018

  8. [7]

    Semi-supervised condition monitoring and visual- ization of fused magnesium furnace,

    S. Lu and Y . Wen, “Semi-supervised condition monitoring and visual- ization of fused magnesium furnace,” IEEE Transactions on Automation Science and Engineering , vol. 19, no. 4, pp. 3471–3482, 2022. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MAY 2024 15 TABLE II: Performance comparison of different randomized methods Model Testing Ra/% P A/MB Trainin...

    work page 2022

Show all 37 references
  1. [8]

    Abnormal condition diagnosis through deep learning of image sequences for fused magnesium furnaces,

    G. Wu, Q. Liu, and T. Chai, “Abnormal condition diagnosis through deep learning of image sequences for fused magnesium furnaces,” Acta Automatica Sinica, vol. 45, no. 8, pp. 1475–1485, 2019

    2019

  2. [9]

    Operating performance assessment based on multi-source heterogeneous information with deep learning for smelting process of electro-fused magnesium furnace,

    K. Bu, L. Yan, and F. Wang, “Operating performance assessment based on multi-source heterogeneous information with deep learning for smelting process of electro-fused magnesium furnace,”ISA Transactions, vol. 128, pp. 357–371, 2022

    2022

  3. [10]

    What does a model really look at?: Extracting model-oriented concepts for explaining deep neural networks,

    S. Kim and D. Chae, “What does a model really look at?: Extracting model-oriented concepts for explaining deep neural networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 46, no. 7, pp. 4612–4624, 2024

    2024

  4. [11]

    Evaluating the visualization of what a deep neural network has learned,

    W. Samek, A. Binder, G. Montavon, S. Lapuschkin, and K. M ¨uller, “Evaluating the visualization of what a deep neural network has learned,” IEEE Transactions on Neural Networks and Learning Systems , vol. 28, no. 11, pp. 2660–2673, 2017

    2017

  5. [12]

    Understanding and accelerating neural architecture search with training-free and theory-grounded metrics,

    W. Chen, X. Gong, J. Wu, Y . Wei, H. Shi, Z. Yan, Y . Yang, and Z. Wang, “Understanding and accelerating neural architecture search with training-free and theory-grounded metrics,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 46, no. 2, pp. 749–763, 2024

    2024

  6. [13]

    A multi-level interpretable sleep stage scoring system by infusing experts’knowledge into a deep network architecture,

    H. Niknazar and S. C. Mednick, “A multi-level interpretable sleep stage scoring system by infusing experts’knowledge into a deep network architecture,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 46, no. 7, pp. 5044–5061, 2024

    2024

  7. [14]

    A generic graph-based neural architecture encoding scheme with multifaceted information,

    X. Ning, Y . Zheng, Z. Zhou, T. Zhao, H. Yang, and Y . Wang, “A generic graph-based neural architecture encoding scheme with multifaceted information,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 7, pp. 7955–7969, 2023

    2023

  8. [15]

    BlockQNN: Efficient block-wise neural network architecture genera- tion,

    Z. Zhong, Z. Yang, B. Deng, J. Yan, W. Wu, J. Shao, and C. Liu, “BlockQNN: Efficient block-wise neural network architecture genera- tion,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 43, no. 7, pp. 2314–2328, 2021

    2021

  9. [16]

    Functional-link net computing: Theory, system architecture, and functionalities,

    Y . H. Pao and Y . Takefuji, “Functional-link net computing: Theory, system architecture, and functionalities,” Computer, vol. 25, no. 5, pp. 76–79, 1992

    1992

  10. [17]

    2-D stochastic configuration networks for image data analytics,

    M. Li and D. Wang, “2-D stochastic configuration networks for image data analytics,” IEEE Transactions on Cybernetics , vol. 51, no. 1, pp. 359–372, 2021

    2021

  11. [18]

    Deep stochastic configuration networks with uni- versal approximation property,

    D. Wang and M. Li, “Deep stochastic configuration networks with uni- versal approximation property,” in 2018 International Joint Conference on Neural Networks (IJCNN) , 2018, pp. 1–8

    2018

  12. [19]

    Online self-learning stochastic configura- tion networks for nonstationary data stream analysis,

    K. Li, J. Qiao, and D. Wang, “Online self-learning stochastic configura- tion networks for nonstationary data stream analysis,”IEEE Transactions on Industrial Informatics , vol. 20, no. 3, pp. 3222–3231, 2024

    2024

  13. [20]

    Predicting particle size of copper ore grinding with stochastic configuration networks,

    D. Wang, P. Tian, W. Dai, and G. Yu, “Predicting particle size of copper ore grinding with stochastic configuration networks,” IEEE Transactions on Industrial Informatics , vol. 20, no. 11, pp. 12 969–12 978, 2024

    2024

  14. [21]

    Fuzzy stochastic configuration networks for nonlinear system modeling,

    K. Li, J. Qiao, and D. Wang, “Fuzzy stochastic configuration networks for nonlinear system modeling,” IEEE Transactions on Fuzzy Systems , vol. 32, no. 3, pp. 948–957, 2024

    2024

  15. [22]

    Ensemble stochastic configuration networks for estimating prediction intervals: A simultaneous robust training algorithm and its application,

    J. Lu, J. Ding, X. Dai, and T. Chai, “Ensemble stochastic configuration networks for estimating prediction intervals: A simultaneous robust training algorithm and its application,” IEEE Transactions on Neural Networks and Learning Systems , vol. 31, no. 12, pp. 5426–5440, 2020

    2020

  16. [23]

    Hierarchical-bayesian-based sparse stochastic configuration networks for construction of prediction inter- vals,

    J. Lu, J. Ding, C. Liu, and T. Chai, “Hierarchical-bayesian-based sparse stochastic configuration networks for construction of prediction inter- vals,” IEEE Transactions on Neural Networks and Learning Systems , vol. 33, no. 8, pp. 3560–3571, 2022

    2022

  17. [24]

    Deep learning,

    Y . LeCun, Y . Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015

    2015

  18. [26]

    1xN pattern for pruning convolutional neural networks,

    M. Lin, Y . Zhang, Y . Li, B. Chen, F. Chao, M. Wang, S. Li, Y . Tian, and R. Ji, “1xN pattern for pruning convolutional neural networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 4, pp. 3999–4008, 2023

    2023

  19. [27]

    Discrimination-aware network pruning for deep model compression,

    J. Liu, B. Zhuang, Z. Zhuang, Y . Guo, J. Huang, J. Zhu, and M. Tan, “Discrimination-aware network pruning for deep model compression,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 44, no. 8, pp. 4035–4051, 2022

    2022

  20. [28]

    Multi-objective evolutionary architectural pruning of deep convolutional neural networks with weights inheritance,

    K. Chung, C. Lee, Y . Tsang, C. Wu, and A. Asadipour, “Multi-objective evolutionary architectural pruning of deep convolutional neural networks with weights inheritance,” Information Sciences , vol. 685, no. 121265, pp. 1–18, 2024

    2024

  21. [29]

    Stochastic configuration networks: Fundamentals and algorithms,

    D. Wang and M. Li, “Stochastic configuration networks: Fundamentals and algorithms,” IEEE Transactions on Cybernetics, vol. 47, no. 10, pp. 3466–3479, 2017

    2017

  22. [30]

    Stochastic configuration networks for self-blast state recognition of glass insulators with adaptive depth and multi-scale representation,

    W. Li, Q. Zhang, D. Wang, W. Sun, and Q. Li, “Stochastic configuration networks for self-blast state recognition of glass insulators with adaptive depth and multi-scale representation,” Information Sciences , vol. 604, pp. 61–79, 2022

    2022

  23. [31]

    On the calculation of convolutions with gaussian kernels,

    M. L. Bailon and D. J. Horntrop, “On the calculation of convolutions with gaussian kernels,” Applied Mathematics and Computation, vol. 176, no. 1, pp. 383–387, 2006

    2006

  24. [32]

    SlimConv: Reducing channel redundancy in convolutional neural networks by features recombining,

    J. Qiu, C. Chen, S. Liu, H. Y . Zhang, and B. Zeng, “SlimConv: Reducing channel redundancy in convolutional neural networks by features recombining,” IEEE Transactions on Image Processing, vol. 30, pp. 6434–6445, 2021

    2021

  25. [33]

    Human-level control through deep reinforcement learning,

    V . Mnih, K. Kavukcuoglu, D. Silver, A. A. Rusu, J. Veness, M. G. Bellemare et al. , “Human-level control through deep reinforcement learning,” Nature, vol. 518, no. 7540, pp. 529–533, 2015

    2015

  26. [34]

    2D convolutional stochastic configuration net- works,

    J. Li and D. Wang, “2D convolutional stochastic configuration net- works,” Knowledge-Based Systems , vol. 300, no. 112249, pp. 1–12, 2024

    2024

  27. [35]

    A survey of convolutional neural networks: Analysis, applications, and prospects,

    Z. Li, F. Liu, W. Yang, S. Peng, and J. Zhou, “A survey of convolutional neural networks: Analysis, applications, and prospects,” IEEE Transac- tions on Neural Networks and Learning Systems , vol. 33, no. 12, pp. 6999–7019, 2022

    2022

  28. [36]

    Stochastic configuration machines for industrial artificial intelligence,

    D. Wang and M. J. Felicetti, “Stochastic configuration machines for industrial artificial intelligence,” 2023, arXiv:2308.13570. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MAY 2024 16 Weitao Li received the Ph.D. degree is industrial automation from Northeastern University,...

    2023 arXiv

  29. [2018]

    in Control Science and Engineering as a Master’s-Ph.D

    She is currently pursuing a Ph.D. in Control Science and Engineering as a Master’s-Ph.D. com- bined student at the School of Electrical Engineering and Automation, Hefei University of Technology. Her research interest covers deep learning, image processing, and industrial arti...

    1995

Pith tools

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