Sure Matt, My original suggestion was intended to just get rid of the "old Temporal Pooler" problem. Until we actually decide on what to change, I suggest we adopt the notion that "TP means Temporal Predictor" for now.
Regards, Fergal On Wed, May 14, 2014 at 5:50 PM, Matthew Taylor <[email protected]> wrote: > Hey guys, before we start making any decisions, I want to make sure we > involve Jeff in this discussion, and he is out of the office for the > next 2+ weeks. I suggest we table this for now and return to it when > Jeff can give his opinion (and he *will* have an opinion). > > Regards, > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > > > On Wed, May 14, 2014 at 7:47 AM, Fergal Byrne > <[email protected]> wrote: > > +1 Hideaki, good points. > > > > I feel using redundant terms across the main functionalities (Memory or > > Pooler) is a waste of characters and obscures the meaning. Also, using > > multiple words hides what is the most important thing. For example: > > > > SP or SpatialPooler would be better as Recogniser or Patterner or > > Representer (1 word), or SpatialRepresenter etc (2 words) > > (old) TP or TemporalPooler "" "" Predictor or ContextPredictor > > (new) Temporal Pooler might be Stabiliser or StableSequencer etc (or just > > Temporal Pooler) > > > > I like the use of Context in old TP, it does add meaning. > > > > > > > > On Wed, May 14, 2014 at 1:54 PM, Hideaki Suzuki <[email protected]> > wrote: > >> > >> Hi, > >> > >> +1 for cleaner terms. Cells4 is confusing for me, too. > >> > >> The core idea of the previous TP is to memorize, IIUC, a chain of > >> "contexts" rather than a mere sequence of inputs, so we can say a 2 is > >> followed by 1 in 1-4-1-4-2 (=sqrt two) while the same 2 is followed by > >> 0 in 1-7-3-2 (=sqrt three). Cell-column structure (or a mini column?) > >> is the enabler. > >> It would be nice if the new name could express that, too. :-) > >> > >> If we can make the new set and the old set of terms disjoint, the > >> confusion may be less... > >> > >> old terms: > >> spacial pooler(SP), temporal pooler(TP) > >> new term: > >> spacial pooling memory(SPM), context chaining memory(CCM), > >> temporal pooling memory(TPM) > >> or spacial pattern pooler (SPP), context pooler (CP), temporal > >> pattern pooler (TPP) > >> > >> Only an idea. > >> > >> _______________________________________________ > >> 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 > > > > _______________________________________________ > 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
