This book, written by a leader in neural network theory in Russia, uses mathematical methods in combination with complexity theory, nonlinear dynamics and optimization. It details more than 40 years of Soviet and Russian neural network research and presents a systematized methodology of neural networks synthesis. The theory is expansive: covering not just traditional topics such as network architecture but also neural continua in function spaces as well.It is important to analyze intermediate modes of the transfer from the problem of learning to the problem of self-learning and vice versa in the algorithm block- scheme represented in Fig. 5.3. The transfer must be performed by the change of theanbsp;...

Title | : | Neural Networks Theory |

Author | : | Alexander I. Galushkin |

Publisher | : | Springer Science & Business Media - 2007-10-29 |

Continue