molander@ae.chalmers.se (soren molander) (05/29/89)
I can't find the ART-code that Mr. Jagath Samarubandu was kind enough
to mail to this newsgroup. This could be due to the fact that our local
net does not subscribe to comp.sources.misc, or other to me unknown reasons.
Could I be so bold as to ask whether the ART-code could be mailed
to comp.sources.misc once again?
While I'm at it, I might mention that I've read - parts of - Minsky and
Papert's book "The Perceptron". It seems to me that many of their claims
seem quite reasonable, and that their methods seem to be generally applicable
to most networks of varying architecture. One of the things I liked was
that they stressed the necessity to treat specific examples as opposed
to the approach of formulating "The General Theory of Perceptrons".
I also liked the stringency of the treatment of the subject itself;
I sometimes get the feel, when reading some of the claims in this
newsgroup, that the lack of a theoretical backbone is a hindrance when
putting forward new ideas.
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
S. Molander, dept. of Applied Electronics | E-mail: molander@ae.chalmers.se
Chalmers Insititue of Technology |
412 96, Sweden. |
=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
"History teaches us that we don't learn
from history" ( With apologies to Sting)