TR available from the neuroprose archive

[Christian Omlin]

The following paper has been placed in the Neuroprose archive.
Comments and questions are encouraged.

*******************************************************************

         --------------------------------------------
            TRAINING SECOND-ORDER RECURRENT NEURAL
                    NETWORKS USING HINTS
         --------------------------------------------


    C.W. Omlin*                       C.L. Giles     
    Computer Science Department       *NEC Research Institute     
    Rensselaer Polytechnic Institute  4 Independence Way         
    Troy, N.Y. 12180                  Princeton, N.J. 08540      
    omlinc at turing.cs.rpi.edu          giles at research.nj.nec.com  



                         Abstract
                         --------

We investigate a method for inserting rules into discrete-time
second-order recurrent neural networks which are trained to
recognize regular languages. The rules defining regular languages
can be expressed in the form of transitions in the corresponding
deterministic finite-state automaton. Inserting these rules as hints
into networks with second-order connections is straight-forward.
Our simulations results show that even weak hints seem to improve
the convergence time by an order of magnitude.


(To be published in Machine Learning: Proceedings of the Ninth
International Conference (ML92),D. Sleeman and P. Edwards (eds.), 
Morgan Kaufmann, San Mateo, CA 1992).


********************************************************************


Filename: omlin.hints.ps.Z

----------------------------------------------------------------
FTP INSTRUCTIONS

     unix% ftp archive.cis.ohio-state.edu (or 128.146.8.52)
     Name: anonymous
     Password: anything
     ftp> cd pub/neuroprose
     ftp> binary
     ftp> get omlin.hints.ps.Z
     ftp> bye
     unix% zcat omlin.hints.ps.Z | lpr
(or whatever *you* do to print a compressed PostScript file)
----------------------------------------------------------------


----------------------------------------------------------------------------
Christian W. Omlin		   Troy, NY 12180 USA			
Computer Science Department        Phone: (518) 276-2930 Fax: (518) 276-4033   
Amos Eaton 119			   E-mail: omlinc at turing.cs.rpi.edu             
Rensselaer Polytechnic Institute           omlinc at research.nj.nec.com
----------------------------------------------------------------------------