Why is NuPIC's SP not deterministic? Optimization reasons?
Is there any insight that can be shared to show why learning creates a
different SDR when learning occurs?
Am I correct inthat the SP needs to create sparcity so if other input come in
which creates a similar SDR then those two SDRs need to have more 'distance'
put between each other?
Kind regards
Stewart
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
_______________________________________________
nupic mailing list
[email protected]
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
