Thank Dr. Byrne and Dr. Hawkins for your enlightening response.. Looking
forward to more discussions.
--
Sai Kumar Dwivedi
___
nupic mailing list
nupic@lists.numenta.org
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
nto SDRs. As long as the output is a sparse array
of semantic bits that capture the important parts of the problem space this
should work.
Jeff
From: nupic [mailto:nupic-boun...@lists.numenta.org] On Behalf Of Fergal
Byrne
Sent: Tuesday, July 09, 2013 10:54 AM
To: NuPIC general mailing
Hi Sai,
All inputs to the neocortex are effectively just on-off signals as well as
scalar data (ie brightness, sound intensity, pressure etc) which NuPIC/Grok
encodes into binary data.
It's really important to understand that the binary data for the CLA is
completely different from the binary
Hello NuPIC Community,
I went through the book "On Intelligence" and also through the white paper
of NUMENTA. Something is troubling me lately.
Instead of having binary inputs ( me talking primarily for object
recognition ) we can have curves as input to the system.
*Advantage*
- Noise such
