Corpus ID: 7986092

Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units

@article{Shang2016UnderstandingAI,
  title={Understanding and Improving Convolutional Neural Networks via Concatenated Rectified Linear Units},
  author={W. Shang and Kihyuk Sohn and D. Almeida and H. Lee},
  journal={ArXiv},
  year={2016},
  volume={abs/1603.05201}
}
  • W. Shang, Kihyuk Sohn, +1 author H. Lee
  • Published 2016
  • Computer Science, Mathematics
  • ArXiv
  • Recently, convolutional neural networks (CNNs) have been used as a powerful tool to solve many problems of machine learning and computer vision. In this paper, we aim to provide insight on the property of convolutional neural networks, as well as a generic method to improve the performance of many CNN architectures. Specifically, we first examine existing CNN models and observe an intriguing property that the filters in the lower layers form pairs (i.e., filters with opposite phase). Inspired… CONTINUE READING
    268 Citations
    Doubly Convolutional Neural Networks
    • 39
    • PDF
    Max-min convolutional neural networks for image classification
    • 33
    • PDF
    A Survey of Model Compression and Acceleration for Deep Neural Networks
    • 348
    • PDF
    RePr: Improved Training of Convolutional Filters
    • 21
    • PDF
    Deep convolutional neural networks compression method based on linear representation of kernels
    • 2
    Convolutional Neural Networks
    • 1
    • PDF

    References

    SHOWING 1-10 OF 39 REFERENCES
    Striving for Simplicity: The All Convolutional Net
    • 2,255
    • Highly Influential
    • PDF
    Max-min convolutional neural networks for image classification
    • 33
    • PDF
    Recurrent convolutional neural network for object recognition
    • M. Liang, Xiaolin Hu
    • Computer Science
    • 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2015
    • 587
    • PDF
    Network In Network
    • 3,119
    • Highly Influential
    • PDF
    ImageNet classification with deep convolutional neural networks
    • 55,155
    • Highly Influential
    • PDF
    Spectral Representations for Convolutional Neural Networks
    • 157
    • PDF
    Understanding deep image representations by inverting them
    • 1,123
    • PDF
    Inverting Convolutional Networks with Convolutional Networks
    • 91
    • PDF
    Very Deep Convolutional Networks for Large-Scale Image Recognition
    • 41,287
    • PDF