Connectionists: The SP theory of intelligence: distinctive features and advantages

[Gerry Wolff]

As some people will know, I have been engaged in a long-term project to
simplify and integrate observations and concepts across artificial
intelligence, mainstream computing, mathematics, and human perception and
cognition (http://www.cognitionresearch.org/sp.htm).

I’ve been reluctant to make comparisons with other research in AI and
related areas, mainly because of differences in what people are trying to
achieve. But now I have drafted a paper, intended for publication, that may
be downloaded via this link:

The SP theory of intelligence: distinctive features and advantages
<http://www.cognitionresearch.org/papers/overview/spalt.pdf> (PDF,
http://bit.ly/1WDRbJv).

As its name suggests, the paper describes what distinguishes the SP theory
from other AI-related research, and why, in many cases, the SP theory has
advantages compared with other approaches to AI. One of the biggest
sections is about advantages of the SP system compared with deep learning
in neural networks.

I’m aware that there are likely to be objections to some of the things I
have said. But this is only a draft paper and I would very much like to
hear objections, or any other comments or suggestions, so that I can make
corrections and changes to the paper, as necessary.

If you would be willing to look at the paper and send me your comments or
suggestions, please let me know fairly soon, in advance of when you send
the comments. I will of course acknowledge all contributions of that kind.

Many thanks,

Gerry Wolff

--

Dr Gerry Wolff PhD CEng MIEEE
Cognition Research

jgw [AT] cognitionresearch.org, +44 (0) 1248 712962, +44 (0) 7746 290775,
Skype: gerry.wolff, Web: www.cognitionresearch.org .
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20150819/e08e5922/attachment.html>