Training Robust Deep Neural Networks via Adversarial Noise Propagation.
Aishan LiuXianglong LiuChongzhi ZhangHang YuQiang LiuJunfeng HePublished in: CoRR (2019)
Keyphrases
- neural network
- training process
- image noise
- noisy environments
- training algorithm
- feedforward neural networks
- feed forward neural networks
- pattern recognition
- genetic algorithm
- multi layer perceptron
- greater robustness
- deep architectures
- feed forward
- salt pepper
- backpropagation algorithm
- random noise
- recurrent neural networks
- noise reduction
- multi agent
- training set
- estimation error
- registration errors
- machine learning
- learning algorithm
- training and testing data
- error back propagation
- label noise
- recurrent networks
- back propagation
- deep learning
- missing data
- noisy data
- signal to noise ratio
- training phase
- robust estimation
- neural nets
- noise level