From: Df.Ning@gmail.com   
      
   I used basic particle swarm to train the three-layer BP network   
   consist of one input layers, one  hidden layer and one output layer,   
   and the hidden layer has 15 neurons, so the dimension of PSO was 46 of   
   which 15 input-hidden weights, 15 hidden-output weights, 15 hidden   
   layer threshold value and 1 output layer threshold value.   
   There were two parameters in basic particle swarm as following: the   
   social weight factor "c2", the individual weight factor "c1",  the max   
   epoch of PSO was 10000.  the training data were  fault data of the   
   actuator of some control system.   
   when I used the fault data to train BP network through PSO, after   
   10000 epochs run, the result that pso had got was not unperfect, not   
   better that the result through back propagation train.   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)