Could anyone explain what "Holographic Neural Net" is? I only have a
paper by AND corp. which is the original inventor of the algorithm, but
it is so hard to read. Any new developent/papers on this subject?
Thank you.
Kevin
The contents of this message express only the sender's opinion.
This message does not necessarily reflect the policy or views of
my employer, Merck & Co., Inc. All responsibility for the statements
made in this Usenet posting resides solely and completely with the
sender.
I think the 'holographic neural network' was originally devised
by Sutherland (Sunderland?), whose paper AND distributed with
their marketing literature. I got lost trying to understand that
paper- all I could make out was that the holographic neural
network was supposed to perform some sort of transformation of
the data to a complex space where patterns could be more easily
stored. On-line, someone told me that the original idea was
actually a good one, but the devil was in the details thus this
method simply didn't work and (supposedly) that the original
inventor had given up on it. The demonstration software which
AND distributed tested the holographic neural network on the
training data, which is a bad way, statistically speaking, of
testing a model. I think AND has since stopped selling the Hnet
software.
--
Will Dwinnell
Interesting follow-up. I remember reading some advertising literature
years ago -- the claims were phenomenal (unbelievable??). So what else is
new?
--
Allan Kaminsky
Remove *NOSPAM* from return address to reply
One correction: I think Itanis (in Pittsburgh, I believe) was the
distributor of Hnet in the United States, and it was they who
stopped selling it. I was not able to prove nor disprove the
Hnet marketing claims. The demonstration software which I
received (as I said) tested on the training data, which
demonstrates almost nothing of interest. When Greg Itani (sp?),
who once visited this forum was challenged on these claims, his
responses were, from a technical perspective, less than
satisfactory.
--
Will Dwinnell
AND gave a big, fancy presentation at last year's JCIS conference
on their Quantum Holographic Networks. I'm not much good with
complex numbers, but as near as I could make out, the properties
of their method were virtually the same as for linear models,
just harder to understand.
--
Warren S. Sarle SAS Institute Inc. The opinions expressed here
sas...@unx.sas.com SAS Campus Drive are mine and not necessarily
(919) 677-8000 Cary, NC 27513, USA those of SAS Institute.
http://www.zdnet.com/pcmag/news/trends/t970929a.htm
Check this out !!!
Kevin Maro wrote:
> Hi,
>
> Could anyone explain what "Holographic Neural Net" is? I only have a
> paper by AND corp. which is the original inventor of the algorithm, but
> it is so hard to read. Any new developent/papers on this subject?
>