Pith. sign in

REVIEW 4 major objections 7 minor 30 references

Comprehensive and Comparative Analysis between Transfer Learning and Custom Built VGG and CNN-SVM Models for Wildfire Detection

T0 review · 4 major / 7 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Pretrained VGG-16 beats every from-scratch model in wildfire detection, with only three errors on 550 test images.

desk verdict A clear but routine comparison of pretrained vs scratch CNNs for wildfire detection; the numbers are plausible but the design confounds transfer learning with architecture depth, so the central claim overreaches. read the letter →

arxiv 2411.08171 v1 pith:XVK2FE46 submitted 2024-11-12 cs.CV cs.AI

Aditya V. Jonnalagadda , Hashim A. Hashim , Andrew Harris This is my paper · ORCID
classification cs.CVcs.AI
keywords wildfiredetectiontransferlearningVGG-16VGG-19ResNet101customCNNCNN-SVMimageclassification
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 sets out to determine whether transfer learning outperforms training from scratch for binary wildfire detection. It compares three custom-built models (VGG-7, VGG-10, and CNN-SVM) with three ImageNet-pretrained models (VGG-16, VGG-19, and ResNet101) on a mixed internet-plus-FLAME dataset of roughly 4,500 images. Its central claim is that the transfer-learning models are more accurate and produce fewer false alarms and missed fires, with VGG-16 reaching 99.45% test accuracy, two false positives, and one false negative on a 550-image test set. The authors argue that pretrained models generalize better to real-world distractors, while custom models remain attractive mainly for their faster inference in real-time drone use.

What carries the argument

The load-bearing mechanism is frozen-feature transfer learning. Lower convolutional layers of VGG-16 and VGG-19, pretrained on ImageNet's 1,000 classes, are kept untrainable and act as generic extractors of edges, textures, and scene structure; only the final dense layer is collapsed to a single linear node for binary fire/non-fire output. Against this, VGG-7, VGG-10, and CNN-SVM learn every weight from the custom dataset alone, with CNN-SVM using an SVM as the final classifier. The paper's argument is that the pretrained layers supply general visual knowledge that a small wildfire dataset cannot teach from scratch, which is why the VGG transfer models post lower false-positive and false-negative rates.

What would settle it

Run a perceptual-hash duplicate check between the 550 test images and the training set, then re-evaluate all six models on an independent public wildfire dataset; if VGG-16's false positives and false negatives rise toward the custom models' levels, or if a from-scratch model matches it, the claimed transfer-learning advantage is a property of the benchmark rather than the method.

Watch

Extended reading notes

Core claim

On the paper's own numbers, the central discovery is that freezing an ImageNet-pretrained VGG-16 backbone and replacing its classifier with a single linear output node yields 99.45% test accuracy, 99.69% recall, and 99.38% precision, with two false positives and one false negative out of 550 test images. VGG-19 follows at 99.09% accuracy, and the best custom model, CNN-SVM, reaches 96.91% accuracy with 98.46% recall. ResNet101, despite being pretrained, falls to 92.73% accuracy, so the paper's conclusion is not that pretraining always helps, but that transfer learning with VGG features is particularly well suited to this fire/no-fire task. The authors further infer from the results that custom from-scratch models are more prone to false alarms from fire-colored objects and are better reserved for speed-critical real-time deployment.

Load-bearing premise

The comparison depends on the custom fire/non-fire image set being a clean, representative sample with no duplicated or near-duplicated images across the training and test splits, so that the near-perfect accuracies reflect true generalization rather than an easy or contaminated benchmark.

Editorial extensions

If this is right

  • For wildfire detection with limited labeled data, pretrained VGG-16 or VGG-19 should be the default starting point over from-scratch CNNs on accuracy grounds.
  • A 0.31% false-negative rate on VGG-16 means roughly one missed fire per 300 fire images, a level the paper argues is suitable for early-warning systems.
  • Custom shallow models trade a few percentage points of accuracy for speed, making them the paper's recommended option for real-time drone processing.
  • ResNet101's weaker 92.73% accuracy shows that pretraining alone does not guarantee transfer success; architecture choice matters in this domain.

Reading between the lines

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

  • The paper does not test its own explanation that pretrained models resist fire-colored distractors; a dedicated test set of autumn leaves, sunsets, and watchtowers would turn that explanation into a checkable claim.
  • Because the compared models differ in both pretraining and architecture depth, the accuracy gap is not a pure measure of transfer learning; pretraining the same VGG-7 and VGG-10 architectures on ImageNet would isolate the contribution of pretraining.
  • The near-ceiling accuracies on a small private dataset suggest the benchmark may be easy; evaluating all six models on an independent public wildfire dataset with identical preprocessing would show whether the transfer-learning advantage persists.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 7 minor

Summary. The paper compares three custom-built models (VGG-7, VGG-10, and CNN-SVM) trained from scratch with three pretrained models (VGG-16, VGG-19, and ResNet101) for binary wildfire detection on a custom dataset combining internet images and FLAME. On a 550-image test set, the pretrained VGG-16 achieves the highest accuracy (99.45%) with 2 false positives and 1 false negative, and the paper concludes that transfer learning yields more accurate and more generalizable wildfire detection than custom models. The manuscript includes architecture tables, training/validation curves, and confusion-matrix counts for all six models.

Significance. If the claimed comparison were properly controlled, this would be a useful applied study: it uses a realistic mixed-source dataset, reports full confusion-matrix statistics rather than only aggregate accuracy, and compares six architectures under a common wildfire-detection task. The paper's strengths are the explicit architecture descriptions, the test-set TP/FP/FN/TN counts, and the attention to false-alarm and missed-detection trade-offs. However, the central claim is not established by the experiments as designed because the comparison confounds transfer learning with architecture depth, trainable capacity, and model selection on the same data, and because all results come from single runs without leakage checks or statistical uncertainty. The result would be worth publishing only after a controlled ablation and a more rigorous evaluation protocol.

major comments (4)
  1. [III.B, III.C, Table I] The headline comparison confounds transfer learning with architecture depth and trainable capacity. Custom VGG-7 and VGG-10 train all of their parameters (10,090,865 and 6,650,993) from scratch, while pretrained VGG-16 and VGG-19 freeze the convolutional base and train only 263,169 parameters (Table I). Section III.B further states that the VGG-16 architecture trained without pretrained weights 'proved to be ineffective,' so the authors substituted shallower custom models. Consequently, the observed advantage of VGG-16 (99.45% test accuracy) over VGG-7 (96.54%) could be due to depth, initialization, or trainable capacity rather than to transfer learning per se. A same-architecture ablation (e.g., VGG-16/19 trained from scratch on the same data, or ImageNet-initialized VGG-7/10) is needed to support the conclusion in Section V that pretrained models are more generalized.
  2. [II.A, IV.C, Table III] The evaluation does not establish statistical reliability. Section II.A describes only training and validation splits (3,629 and 385 images), and the test set of 550 images first appears in Section IV.C with no description of how it was selected, whether it overlaps with training or validation, or whether any duplicate or near-duplicate check was performed. All results come from a single run, with no error bars, confidence intervals, or repeated-seed experiments, so the near-perfect accuracies and the small differences between VGG-16 and the custom models cannot be meaningfully assessed. The manuscript should report the split procedure, a leakage check, and repeated-run statistics.
  3. [III.A] The custom models are tuned on the same dataset used for evaluation, which biases the reported test metrics. Section III.A says the CNN-SVM configuration 'was chosen due to its optimal results across test, train, and validation accuracies,' and Section III.B describes 'repetitive trials' for VGG-7 and VGG-10. Because the test set is part of the model-selection loop, the custom models' test numbers are optimistically selected, while the pretrained models are not selected on this criterion. A fixed test set must be held out before any architecture or hyperparameter choices are made.
  4. [Table I, III.C] The paper's description of the pretrained VGG models is internally inconsistent. Section III.C states the output activation was changed from softmax to linear for the binary task, but Table I lists softmax for both VGG-16 and VGG-19. In addition, Table I gives the VGG-16 conv5 filters as 1,024 and the VGG-19 conv2 filters as 64, whereas the non-trainable parameter counts (14,714,688 and 20,024,384) match the standard Keras VGG-16/VGG-19 feature extractors with 512 filters in the last block and 128 filters in VGG-19's second block. Please correct Table I and specify the loss function and thresholding used with the linear output.
minor comments (7)
  1. [II.A] The dataset description should state the total number of unique images, the number of images per class in the test set, and the process used to create the test split; currently the only mention of a 550-image test set is in Section IV.C.
  2. [IV.B] The text says 'as presented in Table I' when reporting ResNet101 validation results, but those results are in Table II; Table I is the architecture table.
  3. [IV.B] The sentence listing models in Table II omits VGG-19 ('VGG-16 and ResNet101'), although Table II includes VGG-19; please correct this for consistency.
  4. [References] Several citations do not support their claims: [24] and [25] are not about diverse landscapes for wildfire detection, and [26] for VGG-16/VGG-19 should be Simonyan and Zisserman's paper rather than the Akhloufi et al. drone survey.
  5. [III.C] The phrase 'all layers, except the top layers (input and output), are set to be untrainable' should be reworded: in the experiments, the convolutional base is frozen and only the dense classifier is trained.
  6. [Figure 3] The subplot axes and legends are too small to read at print size; please enlarge them and add clear title and axis labels for each subplot.
  7. [III.D] The description of the ResNet modification is incomplete: ResNet101's original classification layer has 1,000 units, not 512; please specify which layers were frozen and which were replaced.

Circularity Check

1 steps flagged · score 4.0 of 10

CNN-SVM's reported test performance is fitted on the test set itself, making that column a selection artifact; the broader transfer-learning comparison is confounded by design but not definitionally circular.

  1. fitted input called prediction [Section III.A (Custom Built CNN-SVM) and Table III]
    "This specific configuration, revealed in extensive testing of diverse setups, was chosen due to its optimal results across test, train, and validation accuracies."

    The CNN-SVM architecture was selected explicitly because it gave optimal results on the test set, along with train and validation. Table III then reports the same test set of 550 images as if it were an independent evaluation, listing CNN-SVM's true positives and derived accuracy, recall, and precision. The reported test performance is therefore not an out-of-sample prediction; it is a restatement of the selection criterion used to choose the architecture. This makes the CNN-SVM row in Tables II and III a fitted input renamed as a test result rather than independent evidence.

full rationale

The paper's central claim is an empirical comparison, not a derivation from first principles, so most of the evaluation chain is self-contained: accuracy, precision, recall, and false positive/false negative counts are measured on held-out test images. Self-citations (e.g., [1] for augmentation details and [6]-[14] for drone/perception context) are contextual and do not carry the weight of the transfer-learning conclusion. The one genuine circularity is the CNN-SVM configuration: Section III.A states it was chosen because it performed optimally 'across test, train, and validation accuracies,' and Table III subsequently presents its test-set counts as a result; that test performance is forced by the selection procedure. A separate, non-circular weakness is the design of the comparison itself: Section III.B explains that from-scratch VGG-16/19 were dropped after 'repetitive trials' because they were ineffective, and Section III.C freezes all pretrained convolutional layers (only 263,169 trainable parameters), while custom VGG-7 and VGG-10 train 10.09M and 6.65M parameters from scratch. The comparison therefore varies architecture depth/capacity and initialization simultaneously, so the conclusion that pretrained models are more generalized overreaches the experiments. This is a validity concern, not a circularity, and it is repairable with a same-architecture ablation.

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

The paper's central claim rests on hand-tuned architectures, assumptions about dataset representativeness and split integrity, and standard metric computation. No new entities are introduced. The free parameters and domain assumptions are the main load-bearing elements, and they are not independently verified.

free parameters (4)
  • VGG-7 architecture (layer depths, filter counts, dense units) = 2x64, 2x128 conv; 16-unit dense; 10,090,865 params
    Chosen after extensive testing on the dataset (Section III.A) to optimize validation accuracy; this is a hand-tuned design, not a derived quantity.
  • VGG-10 architecture = 2x64, 2x128, 3x256 conv; 16-unit dense; 6,650,993 params
    Similarly selected through repetitive trials to fit the small dataset.
  • CNN-SVM architecture and L2 regularization = 32, 64, 128 filters; L2 regularizer on output
    Configuration revealed in extensive testing of diverse setups (Section III.A); the L2 strength is not specified.
  • Pretrained model fine-tuning choices = Frozen base layers, single 512-unit dense, linear output
    Manual modifications to VGG-16, VGG-19, and ResNet101 (Sections III.C and III.D) aimed at faster learning; these choices affect the comparison.
assumptions (4)
  • domain assumption ImageNet pretrained weights provide transferable features for wildfire detection
    The entire premise of transfer learning; not independently validated outside the paper's results.
  • domain assumption The custom dataset (internet images plus FLAME) is representative of real wildfire scenarios and free of label noise and duplicates
    Section II.A describes the dataset but provides no leakage checks or label verification.
  • domain assumption Train/validation/test splits are random and i.i.d. with no data leakage
    No seed or split procedure is described; near-perfect accuracies could indicate an easy split.
  • standard math Standard classification metrics are computed correctly from the reported confusion matrices
    The metrics appear consistent with the confusion counts, but no code is provided to verify.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Comprehensive and Comparative Analysis between Transfer Learning and Custom Built VGG and CNN-SVM Models for Wildfire Detection." pith.science (2026). https://pith.science/paper/XVK2FE46

@misc{pith2026241108171,
  author       = {Pith},
  title        = {Pith review of: Comprehensive and Comparative Analysis between Transfer Learning and Custom Built VGG and CNN-SVM Models for Wildfire Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XVK2FE46}},
  note         = {Machine review of arXiv:2411.08171}
}
read the original abstract

Contemporary Artificial Intelligence (AI) and Machine Learning (ML) research places a significant emphasis on transfer learning, showcasing its transformative potential in enhancing model performance across diverse domains. This paper examines the efficiency and effectiveness of transfer learning in the context of wildfire detection. Three purpose-built models -- Visual Geometry Group (VGG)-7, VGG-10, and Convolutional Neural Network (CNN)-Support Vector Machine(SVM) CNN-SVM -- are rigorously compared with three pretrained models -- VGG-16, VGG-19, and Residual Neural Network (ResNet) ResNet101. We trained and evaluated these models using a dataset that captures the complexities of wildfires, incorporating variables such as varying lighting conditions, time of day, and diverse terrains. The objective is to discern how transfer learning performs against models trained from scratch in addressing the intricacies of the wildfire detection problem. By assessing the performance metrics, including accuracy, precision, recall, and F1 score, a comprehensive understanding of the advantages and disadvantages of transfer learning in this specific domain is obtained. This study contributes valuable insights to the ongoing discourse, guiding future directions in AI and ML research. Keywords: Wildfire prediction, deep learning, machine learning fire, detection

Figures

Figures reproduced from arXiv: 2411.08171 by the authors.

Figure 1
Figure 1. Sample of images in the training dataset. These images include synthetic images created by data augmentation techniques [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. (a) Working principle (schematic) of transfer learning [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Comparative analysis and performance measures pre [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 22 canonical work pages

  1. [1]

    SegNet: A segmented deep learning based convolutional neural network approach for drones wildfire detection,

    A. V . Jonnalagadda and H. A. Hashim, “SegNet: A segmented deep learning based convolutional neural network approach for drones wildfire detection,” Remote Sensing Applications: Society and Environment , p. 101181, 2024

    work page 2024

  2. [2]

    A novel custom optimized convolutional neural network for a satellite image by using forest fire detection,

    V . Kalaivani and P. Chanthiya, “A novel custom optimized convolutional neural network for a satellite image by using forest fire detection,” Earth Science Informatics , vol. 15, no. 2, pp. 1285–1295, 2022

    work page 2022

  3. [3]

    Imagenet: A large-scale hierarchical image database,

    J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, “Imagenet: A large-scale hierarchical image database,” in 2009 IEEE conference on computer vision and pattern recognition . Ieee, 2009, pp. 248–255

    2009

  4. [4]

    A comprehensive survey on transfer learning,

    F. Zhuang and et al., “A comprehensive survey on transfer learning,” Proceedings of the IEEE , vol. 109, no. 1, pp. 43–76, 2020

    work page 2020

  5. [5]

    A survey on transfer learning,

    S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Trans- actions on knowledge and data engineering , vol. 22, no. 10, pp. 1345– 1359, 2009

    2009

  6. [6]

    A geometric nonlinear stochastic filter for simultaneous localization and mapping,

    H. A. Hashim, “A geometric nonlinear stochastic filter for simultaneous localization and mapping,” Aerospace Science and Technology, vol. 111, p. 106569, 2021

    2021

  7. [7]

    UWB ranging and IMU data fusion: Overview and nonlinear stochastic filter for inertial navigation,

    H. A. Hashim, A. E. Eltoukhy, and K. G. Vamvoudakis, “UWB ranging and IMU data fusion: Overview and nonlinear stochastic filter for inertial navigation,” IEEE Transactions on Intelligent Transportation Systems , vol. 25, pp. 359–369, 2024

    work page 2024

  8. [8]

    Exponentially stable observer-based controller for VTOL-UA Vs without velocity measurements,

    H. A. Hashim, “Exponentially stable observer-based controller for VTOL-UA Vs without velocity measurements,” International Journal of Control, vol. 96, no. 8, pp. 1946–1960, 2023

    work page 1946

Show all 30 references
  1. [9]

    Deep Reinforcement Learning for sim-to-real policy transfer of VTOL-UA Vs offshore docking opera- tions,

    A. M. Ali, A. Gupta, and H. A. Hashim, “Deep Reinforcement Learning for sim-to-real policy transfer of VTOL-UA Vs offshore docking opera- tions,” Applied Soft Computing , vol. 162, p. 111843, 2024

    2024

  2. [10]

    Quaternion-based adaptive backstepping fast terminal sliding mode control for quadrotor UA Vs with finite time convergence,

    A. Shevidi and H. A. Hashim, “Quaternion-based adaptive backstepping fast terminal sliding mode control for quadrotor UA Vs with finite time convergence,” Results in Engineering , vol. 1, pp. 1–12, 2024

    2024

  3. [11]

    Geometric stochastic filter with guaranteed performance for autonomous navigation based on IMU and feature sensor fusion,

    H. A. Hashim, M. Abouheaf, and M. A. Abido, “Geometric stochastic filter with guaranteed performance for autonomous navigation based on IMU and feature sensor fusion,” Control Engineering Practice, vol. 116, p. 104926, 2021

    2021

  4. [12]

    Observer-based controller for VTOL-UA Vs tracking using direct vision-aided inertial navigation measurements,

    H. A. Hashim, A. E. Eltoukhy, and A. Odry, “Observer-based controller for VTOL-UA Vs tracking using direct vision-aided inertial navigation measurements,” ISA transactions , vol. 137, pp. 133–143, 2023

    2023

  5. [13]

    UA V avionics safety, certification, accidents, redundancy, integrity and reli- ability: A comprehensive review and future trends,

    D. Wanner, H. A. Hashim, S. Srivastava, and A. Steinhauer, “UA V avionics safety, certification, accidents, redundancy, integrity and reli- ability: A comprehensive review and future trends,” Drone Systems and Applications, vol. 1, pp. 1–20, 2024

    2024

  6. [14]

    Electronic warfare cyberattacks, countermeasures and defensive aids of UA V avionics: A survey,

    A. Yu, J. Kolotylo, H. A. Hashim, and P. Rennison, “Electronic warfare cyberattacks, countermeasures and defensive aids of UA V avionics: A survey,” Green Energy and Intelligent Transportation , vol. PP, p. PP, 2024

    2024

  7. [15]

    Transfer learning techniques for medical image anal- ysis: A review,

    P. Kora and et al., “Transfer learning techniques for medical image anal- ysis: A review,” Biocybernetics and Biomedical Engineering , vol. 42, no. 1, pp. 79–107, 2022

    2022

  8. [16]

    Transfer learning-based dynamic multiobjective optimization algorithms,

    M. Jiang and et al., “Transfer learning-based dynamic multiobjective optimization algorithms,” IEEE Transactions on Evolutionary Compu- tation, vol. 22, no. 4, pp. 501–514, 2017

    2017

  9. [17]

    Fine-tuning convolutional neural networks for fine art classification,

    E. Cetinic, T. Lipic, and S. Grgic, “Fine-tuning convolutional neural networks for fine art classification,” Expert Systems with Applications , vol. 114, pp. 107–118, 2018

    2018

  10. [18]

    Machine learning in disaster management: recent developments in methods and applications,

    V . Linardos, M. Drakaki, P. Tzionas, and Y . L. Karnavas, “Machine learning in disaster management: recent developments in methods and applications,” Machine Learning and Knowledge Extraction , vol. 4, no. 2, 2022

    2022

  11. [19]

    Pre-trained language models in biomedical domain: A systematic survey,

    B. Wang and et al., “Pre-trained language models in biomedical domain: A systematic survey,” ACM Computing Surveys, vol. 56, no. 3, pp. 1–52, 2023

    2023

  12. [20]

    A thorough review of models, evaluation metrics, and datasets on image captioning,

    G. Luo and et al., “A thorough review of models, evaluation metrics, and datasets on image captioning,” IET Image Processing, vol. 16, no. 2, pp. 311–332, 2022

    2022

  13. [21]

    Analysis of deep learning methods for early wildfire detection systems,

    A. S. Mahdi and S. A. Mahmood, “Analysis of deep learning methods for early wildfire detection systems,” in 2022 5th International Conference on Engineering Technology and its Applications (IICETA) . IEEE, 2022, pp. 271–276

    2022

  14. [22]

    Transferability in deep learning: A survey,

    J. Jiang, Y . Shu, J. Wang, and M. Long, “Transferability in deep learning: A survey,” arXiv preprint arXiv:2201.05867 , 2022

    2022 arXiv

  15. [23]

    Imagenet: A large-scale hierarchical image database,

    J. Deng and et al., “Imagenet: A large-scale hierarchical image database,” in 2009 IEEE conference on computer vision and pattern recognition . Ieee, 2009, pp. 248–255

    2009

  16. [24]

    Dated: Guidelines for creating synthetic datasets for engineering design applications,

    C. Picard, J. Schiffmann, and F. Ahmed, “Dated: Guidelines for creating synthetic datasets for engineering design applications,” arXiv preprint arXiv:2305.09018, 2023

    2023 arXiv

  17. [25]

    Identity mappings in deep residual networks,

    K. He, X. Zhang, S. Ren, and J. Sun, “Identity mappings in deep residual networks,” in 14th European Conference Computer Vision 2016, Amsterdam, Netherlands. Springer, 2016, pp. 630–645

    2016

  18. [26]

    Unmanned aerial vehi- cles for wildland fires: Sensing, perception, cooperation and assistance,

    M. A. Akhloufi, A. Couturier, and N. A. Castro, “Unmanned aerial vehi- cles for wildland fires: Sensing, perception, cooperation and assistance,” Drones, vol. 5, no. 1, p. 15, 2021

    2021

  19. [27]

    Very deep convolutional networks for large-scale image recognition,

    K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556 , 2014

    2014 arXiv

  20. [28]

    An autoencoder with convolutional neural network for surface defect detection on cast components,

    O. Chamberland, M. Reckzin, and H. A. Hashim, “An autoencoder with convolutional neural network for surface defect detection on cast components,” Journal of Failure Analysis and Prevention , vol. 23, no. 4, pp. 1633–1644, 2023

    2023

  21. [29]

    Ethical issues in research using datasets of illicit origin,

    D. R. Thomas and et al., “Ethical issues in research using datasets of illicit origin,” in Proceedings of the 2017 Internet Measurement Conference, 2017, pp. 445–462

    2017

  22. [30]

    Imagenet classification with deep convolutional neural networks,

    A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Advances in neural informa- tion processing systems , vol. 25, 2012

    2012

Pith tools

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