Robust Large Margin Deep Neural Networks

Keyphrases

• Deep Neural Network 100%
• Large Margin 100%
• Jacobian Matrix 75%
• Generalization Error 50%
• Nonlinearity 25%
• Training Samples 25%
• Spectral Norm 25%
• Regularizer 25%
• ImageNet Dataset 25%
• Weight Normalization 25%
• CIFAR-10 25%
• Feedforward Network 25%
• Generalization Properties 25%
• Classification Margin 25%
• Pooling Layer 25%
• Arbitrary Depth 25%
• Batch Normalization 25%
• MNIST 25%
• Residual Network 25%

Computer Science

• Deep Neural Network 100%
• Jacobian Matrix 75%
• Generalization Error 50%
• Feedforward Network 25%
• Experimental Result 25%
• Training Sample 25%
• Batch Normalization 25%
• Residual Neural Network 25%
• Reparametrization 25%