home *** CD-ROM | disk | FTP | other *** search
/ NetNews Usenet Archive 1992 #30 / NN_1992_30.iso / spool / comp / ai / neuraln / 4622 < prev    next >
Encoding:
Text File  |  1992-12-21  |  722 b   |  20 lines
  1. Newsgroups: comp.ai.neural-nets
  2. Path: sparky!uunet!usc!cs.utexas.edu!torn!watserv2.uwaterloo.ca!sun3.vlsi.uwaterloo.ca!sun14.vlsi.uwaterloo.ca!waleed
  3. From: waleed@sun14.vlsi.uwaterloo.ca (Fakhr M. Waleed)
  4. Subject: From PNN to MLP
  5. Message-ID: <BzH57s.Axv@sun3.vlsi.uwaterloo.ca>
  6. Sender: news@sun3.vlsi.uwaterloo.ca
  7. Organization: University of Waterloo
  8. Date: Fri, 18 Dec 1992 21:33:27 GMT
  9. Lines: 9
  10.  
  11. Hi,
  12. I have been trying to map PNN (Gaussian sum networks) to MLP
  13. at least approximately, I have moderate success, any body
  14. knows any references (except the one: Nearest Neighbor to MLP?)
  15.  
  16. I know that any sigmiod can be seen to implement the class posterior
  17. between two Gaussian classes, any body exploited this idea?
  18.  
  19. Thanks
  20.