Diffy: a Déjà vu-Free Differential Deep Neural Network Accelerator.
Mostafa MahmoudKevin SiuAndreas MoshovosPublished in: MICRO (2018)
Keyphrases
- neural network
- artificial neural networks
- neural network model
- back propagation
- pattern recognition
- evolutionary algorithm
- genetic algorithm
- neural network is trained
- feed forward
- recurrent neural networks
- parallel implementation
- training algorithm
- activation function
- learning rules
- feed forward neural networks
- deep learning
- neural model
- image reconstruction from projections
- multi layer perceptron
- bp neural network
- network architecture
- hidden layer
- neural nets
- database
- multilayer perceptron
- fault diagnosis
- training data
- real time