Fast training of recurrent networks based on the EM algorithm |
| |
Authors: | Sheng Ma Chuanyi Ji |
| |
Affiliation: | Dept. of Electr. Comput. and Syst. Eng., Rensselaer Polytech. Inst., Troy, NY. |
| |
Abstract: | In this work, a probabilistic model is established for recurrent networks. The expectation-maximization (EM) algorithm is then applied to derive a new fast training algorithm for recurrent networks through mean-field approximation. This new algorithm converts training a complicated recurrent network into training an array of individual feedforward neurons. These neurons are then trained via a linear weighted regression algorithm. The training time has been improved by five to 15 times on benchmark problems. |
| |
Keywords: | |
|
|