home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!spool.mu.edu!yale.edu!ira.uka.de!sol.ctr.columbia.edu!The-Star.honeywell.com!umn.edu!deci!young
- From: young@deci.cs.umn.edu (Mike Young)
- Subject: Re: Jordan's recurrent neural network
- Message-ID: <young.727807462@deci>
- Sender: news@news2.cis.umn.edu (Usenet News Administration)
- Nntp-Posting-Host: deci.cs.umn.edu
- Organization: University of Minnesota
- References: <1993Jan22.205104.5685@oucsace.cs.ohiou.edu>
- Date: Sat, 23 Jan 1993 16:44:22 GMT
- Lines: 21
-
- In <1993Jan22.205104.5685@oucsace.cs.ohiou.edu> mpmehta@bobcat.ent.ohiou.edu (Manish P. Mehta ISE Grad) writes:
-
- >What i am looking for is a software developed for Jordan's recurrent neural
- >network. If any one has used or developed a software please mail that to me.
- >Or if you know where I can get it please let me know...
-
- The Tlearn software developed at UCSD's Center for Research in Language
- is flexible enough to create recurrent networks of a type at least similar
- to Jordan's. I've used it to create Elman recurrent nets, but not Jordan's.
- It seems to me that Jordan may have had a learning rate twist somewhere in
- his work, and if so that particular aspect would not be implementable in
- Tlearn. Give it a try.
-
- I obtained it via anonymous ftp from crl.ucsd.edu. Let me go poke around to
- find the exact location...
- pub/neuralnets/tlearn.tar.Z
-
- Good luck.
-
- Mike Young
- young@cs.umn.edu -or- young@turtle.psych.umn.edu
-