Natural gradient learning neural networks for modeling and identification of nonlinear systems with memory
| Autor: | Mohamed Ibnkahla, Benoit Pochon |
|---|---|
| Rok vydání: | 2002 |
| Předmět: |
Artificial neural network
Computer science Time delay neural network business.industry Deep learning Computer Science::Neural and Evolutionary Computation MathematicsofComputing_NUMERICALANALYSIS Least squares Backpropagation Nonlinear system Stochastic gradient descent Recurrent neural network Feedforward neural network Artificial intelligence Types of artificial neural networks Gradient descent Stochastic neural network business Algorithm |
| Zdroj: | ICASSP |
| DOI: | 10.1109/icassp.2002.5743977 |
| Popis: | This paper applies natural gradient (NG) learning neural networks (NNs) for modeling and identification of nonlinear systems with memory. The nonlinear system is comprised of a discrete-time linear filter H followed by a zero-memory nonlinearity g(.). The neural network model is composed of a linear adaptive filter Q and a two-layer nonlinear neural network (NN). It is shown that the NG learning method outperforms the ordinary gradient descent method in terms of convergence speed and mean squared error (MSE) performance. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|