Bridgeout: Stochastic Bridge Regularization for Deep Neural Networks

@article{Khan2018BridgeoutSB,
  title={Bridgeout: Stochastic Bridge Regularization for Deep Neural Networks},
  author={Najeeb Khan and J. Shah and I. Stavness},
  journal={IEEE Access},
  year={2018},
  volume={6},
  pages={42961-42970}
}
A major challenge in training deep neural networks is <italic>overfitting</italic>, i.e. inferior performance on unseen test examples compared to performance on training examples. To reduce overfitting, stochastic regularization methods have shown superior performance compared to deterministic weight penalties on a number of image recognition tasks. Stochastic methods, such as Dropout and Shakeout, in expectation, are equivalent to imposing a ridge and elastic-net penalty on the model… Expand
8 Citations
Pruning Convolutional Filters Using Batch Bridgeout
  • PDF
Adaptive Low-Rank Factorization to regularize shallow and deep neural networks
  • PDF
Filter Bank Regularization of Convolutional Neural Networks
  • PDF
Theory of adaptive SVD regularization for deep neural networks
  • 4
Sparseout: Controlling Sparsity in Deep Networks
  • 1
  • PDF
Instance-Level Embedding Adaptation for Few-Shot Learning
  • 5
  • PDF
Multi-Layered Basis Pursuit Algorithms for Classification of MR Images of Knee ACL Tear

References

SHOWING 1-10 OF 45 REFERENCES
Shakeout: A New Regularized Deep Neural Network Training Scheme
  • 40
  • Highly Influential
Swapout: Learning an ensemble of deep architectures
  • 108
  • PDF
Dropout: a simple way to prevent neural networks from overfitting
  • 21,834
  • Highly Influential
  • PDF
Adaptive dropout for training deep neural networks
  • 235
  • PDF
Adam: A Method for Stochastic Optimization
  • 61,208
  • PDF
ImageNet classification with deep convolutional neural networks
  • 61,159
  • PDF
Improving deep neural networks for LVCSR using rectified linear units and dropout
  • 1,011
  • PDF
Deep Residual Learning for Image Recognition
  • 62,797
  • PDF
Dropout Training as Adaptive Regularization
  • 417
  • Highly Influential
  • PDF
...
1
2
3
4
5
...