mlsanityData to 7 Oct 2026

On filter design in deep convolutional neural network

arXiv 2410.21644 ↗First posted 29 Oct 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

The deep convolutional neural network (DCNN) in computer vision has given promising results. It is widely applied in many areas, from medicine, agriculture, self-driving car, biometric system, and almost all computer vision-based applications. Filters or weights are the critical elements responsible for learning in DCNN. Backpropagation has been the primary learning algorithm for DCNN and provides promising results, but the size and numbers of the filters remain hyper-parameters. Various studies have been done in the last decade on semi-supervised, self-supervised, and unsupervised methods and their properties. The effects of filter initialization, size-shape selection, and the number of filters on learning and optimization have not been investigated in a separate publication to collate all the options. Such attributes are often treated as hyper-parameters and lack mathematical understanding. Computer vision algorithms have many limitations in real-life applications, and understanding the learning process is essential to have some significant improvement. To the best of our knowledge, no separate investigation has been published discussing the filters; this is our primary motivation. This study focuses on arguments for choosing specific physical parameters of filters, initialization, and learning technic over scattered methods. The promising unsupervised approaches have been evaluated. Additionally, the limitations, current challenges, and future scope have been discussed in this paper.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 31 of its 74 references that are in the map. First posted on arXiv 29 Oct 2024.

Builds on

31 of 74 references.

References · 31

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2016Densely Connected Convolutional Networkscited by 6,691
  5. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  6. 2014Going Deeper with Convolutionscited by 4,231
  7. 2016Wide Residual Networkscited by 3,057
  8. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,320
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  10. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,818
  11. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,424
  12. 2018Deep Clustering for Unsupervised Learning of Visual Featurescited by 1,020
19 more
  1. 2017Opening the Black Box of Deep Neural Networks via Informationcited by 494
  2. 2016Resnet in Resnet: Generalizing Residual Architecturescited by 110
  3. 2017On weight initialization in deep neural networkscited by 67
  4. 2018Interpretable Convolutional Neural Networks via Feedforward Designcited by 49
  5. 2020PixelHop++: A Small Successive-Subspace-Learning-Based (SSL-based) Model for Image Classificationcited by 37
  6. 2016Understanding Convolutional Neural Networks with A Mathematical Modelcited by 33
  7. 2020Comparing Different Deep Learning Architectures for Classification of Chest Radiographscited by 24
  8. 2017On Data-Driven Saak Transformcited by 22
  9. 2021E-PixelHop: An Enhanced PixelHop Method for Object Classificationcited by 9
  10. 2021Weight Initialization Techniques for Deep Learning Algorithms in Remote Sensing: Recent Trends and Future Perspectivescited by 8
  11. 2019An Image Clustering Auto-Encoder Based on Predefined Evenly-Distributed Class Centroids and MMD Distancecited by 7
  12. 2021Learning to ignore: rethinking attention in CNNscited by 6
  13. 2021Dendritic Self-Organizing Maps for Continual Learningcited by 5
  14. 2020CSNNs: Unsupervised, Backpropagation-free Convolutional Neural Networks for Representation Learningcited by 4
  15. 2019PixelHop: A Successive Subspace Learning (SSL) Method for Object Classificationcited by 3
  16. 2020Teacher-Class Network: A Neural Network Compression Mechanismcited by 3
  17. 2021Beyond Classification: Knowledge Distillation using Multi-Object Impressionscited by 2
  18. 2021A SOM-based Gradient-Free Deep Learning Method with Convergence Analysiscited by 1
  19. 2021Quality Map Fusion for Adversarial Learningcited by 1

Built on by

Not cited yet.