Training Modern Deep Neural Networks for Memory-Fault Robustness.
Ghouthi Boukli HaceneFrançois Leduc-PrimeauAmal Ben SoussiaVincent GriponFrançois GagnonPublished in: CoRR (2019)
Keyphrases
- neural network
- fault diagnosis
- training process
- training algorithm
- feedforward neural networks
- backpropagation algorithm
- training set
- auto associative
- associative memory
- artificial neural networks
- feed forward neural networks
- error back propagation
- training patterns
- multi layer perceptron
- neural network training
- pattern recognition
- memory size
- test set
- deep architectures
- neural network structure
- neural nets
- multi layer
- random access
- computing power
- neural network model
- deep learning
- self organizing maps
- memory usage
- e learning
- fault detection
- training samples
- fuzzy neural network
- memory space
- data sets
- online learning
- object detection
- computational efficiency
- feed forward
- memory requirements
- main memory
- multiple faults
- training data