From: intrest86@hotmail.com   
      
   Hello everyone,   
      
   If this isn't the appropriate Newsgroup for this post, just let me   
   know.   
      
   Basically I am looking for a way to feed in a long string of symbols,   
   and then feed in another shorter string of symbols and check if the   
   second string is a substring of the first.  However, the second string   
   would be noisy.  In the end, I am really just looking for a confidence   
   value of how likely there is a match.   
      
   I'm not sure if this is possible with RNNs or LSTMs.  I am thinking   
   that it just is impossible for classical RNNs since there could be a   
   long time lag between feeding in the match of the substring in the   
   first string, and feeding in the actual substring.  That is why I am   
   looking more at LSTMs because of their ability to store information for   
   a longer period of time without it decaying away.   
      
   But is it even possible?  The first string could be of significant   
   size, and it doesn't seem like there is anyway it could solve the   
   problem without somehow storing ALL of the large string, which would   
   mean an arbitrarily large number of memory cells.  On the other hand,   
   I'm only looking for a probability, so if an NN can give a reasonable   
   probability, then that would still be success.   
      
   Ok, I am rambling.  Basically, I am new to NNs and am trying to get a   
   feel for their applicability.  I know that they can do pattern   
   matching, but how about subpatterns?   
      
   [ comp.ai is moderated.  To submit, just post and be patient, or if ]   
   [ that fails mail your article to , and ]   
   [ ask your news administrator to fix the problems with your system. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)