发明名称 |
METHODS OF IMPROVED LEARNING IN SIMULTANEOUS RECURRENT NEURAL NETWORKS |
摘要 |
<p>Methods, computer-readable media, and systems are provided for machine learning in a simultaneous recurrent neural network. One embodiment of the invention provides a method including initializing one or more weight in the network, initializing parameters of an extended Kalman filter, setting a Jacobian matrix to an empty matrix, augmenting the Jacobian matrix for each of a plurality of training patterns, adjusting the one or more weights using the extended Kalman filter formulas, and calculating a network output for one or more testing patterns.</p> |
申请公布号 |
WO2010042256(A9) |
申请公布日期 |
2010.06.03 |
申请号 |
WO2009US46036 |
申请日期 |
2009.06.02 |
申请人 |
THE UNIVERSITY OF MEMPHIS RESEARCH FOUNDATION;KOZMA, ROBERT;WERBOS, PAUL, J. |
发明人 |
KOZMA, ROBERT;WERBOS, PAUL, J. |
分类号 |
G06N3/08 |
主分类号 |
G06N3/08 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|