Selected Newsgroup Message
From: "Sergio Navega" <snavega@ibm.net>
Subject: Re: Biologically plausible mechanism for learning NL
Date: 07 Apr 1999 00:00:00 GMT
Message-ID: <370bb689@news3.us.ibm.net>
References: <370B84A0.D1ECF8ED@axcess.co.uk>
X-Notice: should be reported to postmaster@ibm.net
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
X-Complaints-To: postmaster@ibm.net
X-Trace: 7 Apr 1999 19:48:25 GMT, 129.37.182.154
Organization: SilWis
Newsgroups: comp.ai.nat-lang
Karla Parussel wrote in message <370B84A0.D1ECF8ED@axcess.co.uk>...
>Hi!
>
>I'm looking for any decent papers that people can recommend regarding
>the biologically plausible mechanisms that are thought to be employed by
>human infants when learning natural language. It's quite possible I
>might be flamed out of this newsgroup for not being a symbolist but I am
>specifically looking to apply these mechanisms using biologically
>plausible neural networks. (well, it's something to do isnt it? :)
>
>Thanks for any help you can give me :)
>
Karla, you will not get that flaming from me. Although I like symbolicist
methods, there's a lot of good work done in that area from connectionists.
This is a very interesting subject and one that I think is in the core of
the problem of intelligence (language learning by children). Let me
suggest some references:
Elman, Jeffrey, and Bates, E., Johnson, M., Karmiloff-Smith, A.,
Parisi, D., Plunkett, K. (1996) Rethinking Innateness,
A Connectionist Perspective on Development. MIT Press.
This book is a must. A clear and witty attack to the innateness of
language position, Chapter 2 (Why Connectionism?) seems to address what
you're after.
Karmiloff-Smith, Annette (1992) Beyond Modularity. MIT Press.
This book is a significant refurbishment of Piaget's developmental
theories. Chapter 2 (Child as a Linguist) seems to address some of
your concerns.
Brent, Michael R. (1997) Computational Approaches to Language
Acquisition. MIT Press.
McKeown, Margaret G., Curtis, Mary E. (eds) (1987) The Nature of
vocabulary acquisition. Lawrence Erlbaum Associates.
If you search the web you'll find lots of additional materials.
For a great work in connectionist NLP, see Risto Miikkulainen's
homepage:
http://www.cs.utexas.edu/users/risto/index.html
See also James McClelland:
http://www.cnbc.cmu.edu/people/mcclelland.html
There are lots of other interesting links, let me know if you want
some more.
Regards,
Sergio Navega.
Back to Menu of Messages Sergio's Homepage
Any Comments? snavega@attglobal.net