On the cost effectiveness of logarithmic arithmetic for backpropagation training on SIMD processors.
Mark G. ArnoldThomas A. BaileyJerry J. CupalMark D. WinkelPublished in: ICNN (1997)
Keyphrases
- back propagation
- cost effectiveness
- training algorithm
- feed forward neural networks
- feedforward neural networks
- backpropagation algorithm
- parallel algorithm
- cost effective
- multilayer perceptron neural network
- neural network
- artificial neural networks
- parallel processing
- multilayer neural network
- feed forward
- error back propagation
- bp algorithm
- hidden layer
- multilayer perceptron
- neural nets
- bp neural network
- parallel architectures
- backpropagation neural network
- activation function
- processor array
- mesh connected
- levenberg marquardt
- single instruction multiple data
- fuzzy logic
- trained neural network
- training process
- cascade correlation
- e learning
- learning algorithm
- backpropagation neural networks
- parallel implementation
- extreme learning machine
- multi layer perceptron
- neural network ensemble
- learning rate
- recurrent neural networks
- multi layer neural network
- parallel computing
- real time
- multi layer
- low cost
- training set
- training data
- machine learning