home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.ai.neural-nets
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!newsserver.jvnc.net!news.edu.tw!cc.ntnu.edu.tw!news!debbie!ccsun10!u8013536
- From: u8013536@cc.nctu.edu.tw ()
- Subject: Re: recurrent neural network (A Question)
- Message-ID: <1993Jan12.220050.22544@debbie.cc.nctu.edu.tw>
- Sender: usenet@debbie.cc.nctu.edu.tw (News Sender)
- Organization: National Chiao Tung University
- X-Newsreader: TIN [version 1.1 PL6]
- References: <1993Jan12.145007.16724@ghost.dsi.unimi.it>
- Date: Tue, 12 Jan 1993 22:00:50 GMT
- Lines: 31
-
- giovanni cartabia (cartabia@ghost.dsi.unimi.it) wrote:
- : I am a C.S. student wich works on neural nets (boltzmann machines) but
- : I heared somethink about recurrent neural networks. Can anyone suggest
- : me some GOOD papers about the argument?
- :
-
- I need those papers, too. But I am a communication engineering student
- working on my M.S. paper.
-
- I train the recurrent network for the purpose of speech recognize.
- But I can not find any paper that explain the behavior of the network.
- So I have a question: "How do the recurrent network work ?".
-
- In a book I see those words...
-
- (Neurocomputing, Robert Hecht-Nielsen, Addison Wesley, p185)
-
- F : (y(0),x(1),x(2),....,x(n)) -> (y(1),y(2),.....,y(n))
-
- "Thus, we can view the operation of such a spatiotemporal system as
- performing a mapping from a set consisting of the initial system states and a
- set of input values provided over the run to a set consisting of thr output
- states produced by the system over the run."
-
- Could anyone tell me more information about the theory of the recurrent
- network ? Any help is appreciated. And thanks in advance.
-
- liao,
- u8013536@cc.nctu.edu.tw
-
-
-