Jakub, (01)
Thanks for those references. I am firmly convinced that the
discrete symbols of language and logic, although important, must
be supplemented with continuous computational methods. The Liquid
State Machine (LSM) is an interesting contribution to the repertoire
of continuous methods: (02)
http://www.lsm.tugraz.at/learning/framework.html (03)
It illustrates the enormous range of radically different computational
methods that haven't yet been adequately explored. Although nobody
really knows how the brain processes information, I suspect that
something like the LSM is closer to what goes on than either a
conventional digital computer or the proposed quantum computers. (04)
Given the large numbers of computational models that are possible
and the shortage of information about the nature of neurons and
how they operate, either individually or collectively, I suspect
that it will be a long, long time before we get any realistic
neural models of how the brain does whatever it does. (05)
Yet the neural network models of the past 50 years, although
hopelessly unrealistic as models of how the brain works, have
still proved to be useful for many applications. Perceptrons,
for example, were first proposed in 1957, and updated versions
of perceptrons are still widely used today. (06)
I suspect that the LSM and/or things like it will prove to be
similarly fruitful for a new generation of computational systems. (07)
John Sowa (08)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (09)
|