Back to Search
Start Over
Second order back-propagation learning algorithm and its application for neural network
- Source :
- Proceedings of the 3rd World Congress on Intelligent Control and Automation (Cat. No.00EX393).
- Publication Year :
- 2002
- Publisher :
- IEEE, 2002.
-
Abstract
- In this paper, a new second order recursive learning algorithm to multilayer feedforward network is proposed. This algorithm makes not only each layer's errors but also second order derivative information factors backpropagate. And it is proved that it is equivalent to Newton iterative algorithm and has second order convergent speed. New algorithm achieves the recurrence calculation of Newton search directions and the inverse of Hessian matrices. Its calculation complexity corresponds to that of common recursive least squares algorithm. It is stated clearly that this new algorithm is superior to Karayiannis' second order algorithm (1994) according to analysis of their properties.
- Subjects :
- Recursive least squares filter
Hessian matrix
Freivalds' algorithm
Binary GCD algorithm
Wake-sleep algorithm
Artificial neural network
Computer science
Iterative method
Dinic's algorithm
Population-based incremental learning
Feed forward
Out-of-kilter algorithm
Backpropagation
Matrix (mathematics)
symbols.namesake
Ramer–Douglas–Peucker algorithm
symbols
Feedforward neural network
Algorithm design
Suurballe's algorithm
Difference-map algorithm
Newton's method
Algorithm
Second derivative
FSA-Red Algorithm
Subjects
Details
- Database :
- OpenAIRE
- Journal :
- Proceedings of the 3rd World Congress on Intelligent Control and Automation (Cat. No.00EX393)
- Accession number :
- edsair.doi...........a722027a480d2612286cfaea7e101388
- Full Text :
- https://doi.org/10.1109/wcica.2000.863343