Leistritz, Lutz Galicki, Miroslaw Witte, Herbert Kochs, Eberhard <br/><br/>
Initial State Training Procedure Improves Dynamic Recurrent Networks with Time-Dependent Weights 

 - 1513-1518 p. 
<br/><br/><label>Subjects--Topical Terms: </label><br/>Dynamic Neural Networks<br/>Optimal Control<br/>Initial Network States