Neural Network Training with Extended Kalman Filter Using Graphics Processing Unit.
Peter TrebatickýJiri PospichalPublished in: ICANN (2) (2008)
Keyphrases
- neural network training
- extended kalman filter
- graphics processing units
- training algorithm
- general purpose
- neural network
- back propagation
- parallel computing
- parallel implementation
- parallel processing
- computing systems
- kalman filter
- massively parallel
- kalman filtering
- real time
- support vector machine
- floating point
- estimation accuracy
- state estimation
- efficient implementation
- learning rate
- high performance computing
- training process
- multilayer perceptron
- particle filter
- learning algorithm
- artificial intelligence
- rbf neural network
- hidden layer
- support vector
- artificial neural networks
- optimization method
- pairwise
- knn