Hi Manuel, We're having an in-depth discussion about this over on [nupic-theory]. Essentially, the thing we currently call the Temporal Pooler is just predicting the next input, which is not the original meaning of Temporal Pooling. We're currently deciding on the right names for the various pieces, so I won't confuse the matter any further!
Regards, Fergal Byrne On Sun, Jun 8, 2014 at 12:56 PM, Scheele, Manuel < [email protected]> wrote: > Hi All, > > Can someone tell me why we need a new temporal pooler or what is wrong > with the current one? > > I believe to remember reading the phrase "the temporal pooler is unable > to learn stable representations of sequences" somewhere, but can't remember > where. > > Could someone link me to a source that would clarify the problems we > have with the current implementation or give an explanation? > > Regards, > Manuel > > _______________________________________________ > nupic mailing list > [email protected] > http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org > > -- Fergal Byrne, Brenter IT Author, Real Machine Intelligence with Clortex and NuPIC https://leanpub.com/realsmartmachines Speaking on Clortex and HTM/CLA at euroClojure Krakow, June 2014: http://euroclojure.com/2014/ and at LambdaJam Chicago, July 2014: http://www.lambdajam.com http://inbits.com - Better Living through Thoughtful Technology http://ie.linkedin.com/in/fergbyrne/ - https://github.com/fergalbyrne e:[email protected] t:+353 83 4214179 Join the quest for Machine Intelligence at http://numenta.org Formerly of Adnet [email protected] http://www.adnet.ie
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
