John,
Thanks very much for the link to Temple Grandin's TED talk - I enjoyed it.
I agree spatial reasoning is needed for human-level intelligence. The TalaMind architecture is open to future extension and integration of spatial reasoning processes and representations.
I did not have time to address the topic in the scope of my thesis work, so page 16 of thesis says only that it is a topic for future research. In retrospect, I'd like to have spent more time on it, and perhaps can add a section about spatial reasoning to the Postscript, someday.
In re-reading Minsky's 1974 frame paper (from the link you posted to Ontolog-Forum), I was reminded of his discussion of the interplay between visual processing, spatial reasoning, symbolic processing, and linguistic processing. An article earlier this year at:
http://www.kdnuggets.com/2014/06/deep-learning-deep-flaws.html reports that images which are classified correctly by Deep Learning algorithms can be misclassified when small random distortions are applied, although the resulting images appear identical to humans. I suspect this may be because the deep learning algorithms do not perform the kind of symbolic structural processing that Minsky argued is necessary for image processing of spatial scenes...
Phil
> Date: Sat, 13 Dec 2014 13:33:26 -0500
> From: sowa@xxxxxxxxxxx
> To: ontolog-forum@xxxxxxxxxxxxxxxx
> Subject: Re: [ontolog-forum] TalaMind Child architecture
>
> Phil,
>
> Spatial reasoning is the foundation for all reasoning, including
> linguistic. In fact, child thinking begins with spatial thinking.
>
> > Spatial reasoning is another area for future research...
>
> Temple Grandin is an autistic woman who was slow in learning to speak.
> She thinks almost entirely in images, and she has great difficulty
> in understanding abstractions.
>
> I recommend her TED talk about ways of thinking:
>
> http://www.ted.com/talks/temple_grandin_the_world_needs_all_kinds_of_minds
>
> Mathematical notation is based on linguistic patterns, but spatial
> reasoning is fundamental to mathematical reasoning and discovery.
> Formal proofs are based on notation, but they're "housekeeping" for
> publication, not discovery. For discussion, see slides 2 to 10 of
>
> http://www.jfsowa.com/talks/micai.pdf
> Why has AI failed? And how can it succeed?
>
> John
>
> _________________________________________________________________
> Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
> Config Subscr: 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 join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
>