论文信息 - Periodic activation function for fast on-line EKF training and pruning

Periodic activation function for fast on-line EKF training and pruning

In the use of the local extended Kalman filter algorithm for network training and pruning, the authors propose replacing the monotonical increasing sigmoid activation function with a periodic function. Computer simulations show that the rate of convergence is substantially improved and that more insignificant weights can be pruned.

Kwok-Wo Wong | Chi-Sing Leung | Sheng-Jiang Chang

[1] Kwok-Wo Wong,et al. Recursive least squares approach to combining principal and minor component analyses , 1998 .

[2] Lai-Wan Chan,et al. On-line training and pruning for recursive least square algorithms , 1996 .

[3] Kwok-Wo Wong,et al. Recursive algorithms for principal component extraction , 1997 .

[4] Hideaki Sakai,et al. A real-time learning algorithm for a multilayered neural network based on the extended Kalman filter , 1992, IEEE Trans. Signal Process..

[5] Sharad Singhal,et al. Training feed-forward networks with the extended Kalman algorithm , 1989, International Conference on Acoustics, Speech, and Signal Processing,.