HI David, It's because time is not actual data (unless fed in as an encoded input to the SP, and it would lose its identity in that case). CLA learns sequences using transitions (with an atomic gap in between), and there is no concept of time as a scalar.
Regards Fergal Byrne On Fri, Aug 8, 2014 at 6:21 PM, cogmission1 . <[email protected]> wrote: > >3. Naming issues related to "new Temporal Pooling", use of TP in past > discussions, code, and mailing list discussions. This has not been > addressed, and perhaps an online discussion would be better than an > email-based one (which clearly hasn't done the job). Perhaps Office Hours > is not the right forum for this; in which case we could still plan a > Hangouts-based session to debate, discuss and agree on the way forward. > > Re: > > I noticed that the actual time stamp is taken out of the new > TemporalMemory, relegating the new design to one that observes sequence > only - regardless of timing between inputs... Was wondering what the > thinking behind this was? > > David > > > On Fri, Aug 8, 2014 at 12:18 PM, Fergal Byrne <[email protected] > > wrote: > >> Hi Matt, >> >> Possible ideas: >> >> 1. Geospatial encoders. I think a Q&A with Chetan and Jeff would be >> really interesting and informative. >> >> 2. Jeff's new Temporal Pooling theory. A lot of people are still confused >> about lots of this, and I feel the talks so far have been mostly exposition >> by Jeff rather than interactive. >> >> 3. Naming issues related to "new Temporal Pooling", use of TP in past >> discussions, code, and mailing list discussions. This has not been >> addressed, and perhaps an online discussion would be better than an >> email-based one (which clearly hasn't done the job). Perhaps Office Hours >> is not the right forum for this; in which case we could still plan a >> Hangouts-based session to debate, discuss and agree on the way forward. >> >> Regards >> >> Fergal Byrne >> >> >> On Thu, Aug 7, 2014 at 11:46 PM, Matthew Taylor <[email protected]> wrote: >> >>> We're having an office hour on Monday: >>> >>> >>> https://plus.google.com/b/100642636108337517466/events/c8ogadmp4n91rd8r0ve3ah9v41s >>> >>> Does anyone have any ideas about topics we should focus on? >>> >>> Thanks, >>> --------- >>> Matt Taylor >>> OS Community Flag-Bearer >>> Numenta >>> >>> _______________________________________________ >>> 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
