Multi-stream extended Kalman filter training for static and dynamic neural networks
GV Puskorius, LA Feldkamp - 1997 IEEE International …, 1997 - ieeexplore.ieee.org
We discuss a powerful and enabling extension to the class of second order neural network
training methods based on the extended Kalman filter (EKF). EKF training procedures are …
training methods based on the extended Kalman filter (EKF). EKF training procedures are …
Extended Kalman filter neural network training: experimental results and algorithm improvements
F Heimes - SMC'98 Conference Proceedings. 1998 IEEE …, 1998 - ieeexplore.ieee.org
It is well known that the extended Kalman filter (EKF) neural network training algorithm is
superior to the standard backpropagation algorithm. However, there are many variations on …
superior to the standard backpropagation algorithm. However, there are many variations on …
Training neural networks using sequential extended Kalman filtering
ES Plumer - 1995 - osti.gov
Recent work has demonstrated the use of the extended Kalman filter (EKF) as an alternative
to gradient-descent backpropagation when training multi-layer perceptrons. The EKF …
to gradient-descent backpropagation when training multi-layer perceptrons. The EKF …
Comparative analysis of backpropagation and extended Kalman filter in pattern and batch forms for training neural networks
S Li - IJCNN'01. international joint conference on neural …, 2001 - ieeexplore.ieee.org
The extended Kalman filter (EKF) algorithm has been used for training neural networks. Like
the backpropagation (BP) algorithm, the EKF algorithm can be in pattern or batch form. But …
the backpropagation (BP) algorithm, the EKF algorithm can be in pattern or batch form. But …
Roles of learning rates, artificial process noise and square root filtering for extended Kalman filter training
GV Puskorius, LA Feldkamp - IJCNN'99. International Joint …, 1999 - ieeexplore.ieee.org
Singhal and Wu (1989) introduced the extended Kalman filter (EKF) neural network training
algorithm. Since then, many modifications, simplifications and improvements of the EKF …
algorithm. Since then, many modifications, simplifications and improvements of the EKF …
Cascade neural networks with node-decoupled extended Kalman filtering
MC Nechyba, Y Xu - … in Robotics and Automation CIRA'97.' …, 1997 - ieeexplore.ieee.org
Most neural networks used today rely on rigid, fixed-architecture networks and/or slow,
gradient descent-based training algorithms (eg backpropagation). In this paper, we propose …
gradient descent-based training algorithms (eg backpropagation). In this paper, we propose …
Recurrent neural network training by nprKF joint estimation
LA Feldkamp, TM Feldkamp… - Proceedings of the …, 2002 - ieeexplore.ieee.org
We present a method for training recurrent networks with the joint estimation of states and
parameters, using the" derivative-free" formulation for nonlinear Kalman filters by Norgaard …
parameters, using the" derivative-free" formulation for nonlinear Kalman filters by Norgaard …
On-line learning in recurrent neural networks using nonlinear Kalman filters
B Todorovic, M Stankovic… - Proceedings of the 3rd …, 2003 - ieeexplore.ieee.org
The extended Kalman filter has been successfully applied to the feedforward and the
recurrent neural network training. Recently introduced derivative-free filters (unscented …
recurrent neural network training. Recently introduced derivative-free filters (unscented …
[引用][C] Optimal learning rate for training time lagged recurrent neural networks with the extended Kalman filter algorithm
P Sun, K Marko - 1998 IEEE International Joint Conference on …, 1998 - ieeexplore.ieee.org
This paper develops a means to compute an optimal learning rate to improve performance
of time lagged recurrent neural networks (TLFWN) trained with the generalized extended …
of time lagged recurrent neural networks (TLFWN) trained with the generalized extended …
The square root Kalman filter training of recurrent neural networks
P Sun, K Marko - SMC'98 Conference Proceedings. 1998 IEEE …, 1998 - ieeexplore.ieee.org
The conventional Kalman filter suffers from the problem that the covariance matrix may not
remain positive definite. In using the filter to train neural networks, the consequence of this …
remain positive definite. In using the filter to train neural networks, the consequence of this …