Corpus ID: 53035661

DropFilter: Dropout for Convolutions

@article{Chen2018DropFilterDF,
  title={DropFilter: Dropout for Convolutions},
  author={Zhengsu Chen and J. Niu and Q. Tian},
  journal={ArXiv},
  year={2018},
  volume={abs/1810.09849}
}
  • Zhengsu Chen, J. Niu, Q. Tian
  • Published 2018
  • Computer Science
  • ArXiv
  • Using a large number of parameters , deep neural networks have achieved remarkable performance on computer vison and natural language processing tasks. However the networks usually suffer from overfitting by using too much parameters. Dropout is a widely use method to deal with overfitting. Although dropout can significantly regularize densely connected layers in neural networks, it leads to suboptimal results when using for convolutional layers. To track this problem, we propose DropFilter, a… CONTINUE READING
    DropFilterR: A Novel Regularization Method for Learning Convolutional Neural Networks

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 38 REFERENCES
    Very Deep Convolutional Networks for Large-Scale Image Recognition
    38082
    Deep Networks with Stochastic Depth
    808
    Going deeper with convolutions
    19356
    ImageNet Classification with Deep Convolutional Neural Networks
    50698
    ImageNet classification with deep convolutional neural networks
    50698
    Wide Residual Networks
    2210
    Curriculum Dropout
    26