Adversarial Concurrent Training: Optimizing Robustness and Accuracy Trade-off of Deep Neural Networks.
Elahe AraniFahad SarfrazBahram ZonoozPublished in: BMVC (2020)
Keyphrases
- trade off
- neural network
- training process
- bias variance
- training algorithm
- pattern recognition
- feedforward neural networks
- high accuracy
- computational efficiency
- test set
- feed forward neural networks
- multi layer perceptron
- back propagation
- genetic algorithm
- backpropagation algorithm
- neural network model
- classification accuracy
- computational cost
- prediction accuracy
- neural nets
- error rate
- deep architectures
- training samples
- artificial neural networks
- training set
- computational speed
- numerical stability
- training speed
- error back propagation
- computational complexity
- neural network training
- expert systems
- recurrent networks
- training patterns
- generalization ability
- highly accurate
- radial basis function