Hi Fergal,

Thanks for your response.

Regarding the first question:

On Sat, Jan 25, 2014 at 1:12 AM, Fergal Byrne
<[email protected]>wrote:

>
> There was no previous design for Temporal Pooling in the theory. There is
> a bit of the algorithm called Temporal Pooler (TP) which is misnamed, it
> should be called Sequence Recognition or Sequence Prediction (Jeff uses the
> term Sequence Memory so the choice of corrected name is unclear).
>

If I recall the whitepaper correctly, there was indeed a previous design
for Temporal Pooling in the theory. It involved neurons forming connections
to previously-active neurons, stretching farther and farther back in time
as long as there was a stable sequence to learn. I meant to ask
specifically about that idea, compared to the new one Jeff introduced here.


Thank you Fergal, for your time in writing up your thoughts. I would also
love to hear Jeff's take on it as well.

- Chetan
_______________________________________________
nupic mailing list
[email protected]
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org

Reply via email to