Here is my 2 cents: 1a. Back in the vision days they used square connect with hierarchy. So I would start there. 1b. You will have to experiment with this. 2a. Not really sure as I haven't tried it. You will have to experiment. I would probably start with two regions. Square connect in bottom region and try both square and random in the top. And probably just start with somewhere around 2k columns for each with 2% active. 2b. Just need to try things I think. 3a. No need for TP. It won't help with spatial representations. 5. Not totally sure I understand the question but it seems like you are trying to figure out how to get a classification out. You could use the CLA classifier for this. It takes arbitrary SDRs and turns them into a value. In this case your classifier input value would be a category with each character in the data set as a potential category value.
On Mon, Jul 15, 2013 at 9:44 AM, Quinn Liu <[email protected]> wrote: > Hi everyone, > Thank you to everyone that's replied thus far. I greatly appreciate > all the info! In hopes of making it easier for any incoming readers I have > updated the question site at: > http://www.walnutiq.com/#!questions-about-cla/cmr with any answers that > have already been given. > > Best Regards, > Quinn Liu > > [email protected] > > > On Mon, Jul 15, 2013 at 11:44 AM, Fergal Byrne < > [email protected]> wrote: > >> >> Hi Michael, >> >> Handwritten characters are undoubtedly multi-component designs, which >> have evolved to connect with and trigger our ability to learn spatial, >> temporal and hierarchical patterns. We perceive the same characters even >> when loads of things change in fonts, and especially when reading different >> people's handwriting. We can fill in gaps and correct misspellings. So the >> learning and prediction must be several levels deep in hierarchy. >> >> In terms of bottom level mechanics, we use saccades to recognise and >> "delocalise" components such as characters, facial features, etc, in such a >> way as to allow this multi-level recognition (including a hierarchy of >> fixations - for strokes, junctions, topology, characters, letters, words, >> and even sentences). >> >> Speed-readers can saccade to read entire phrases and sentences at a >> time, allowing reading speeds of thousands of words per minute with better >> than 70% comprehension scores. With practice, I've been able to get scores >> in the 1-2000 wpm range. I can also read text in a mirror or upside-down at >> speeds approaching 50-60% of an average reader. These things could only be >> done using big, complex region hierarchies with vast volumes of (normal) >> reading practice. >> >> I would have predicted that a single layer CLA would struggle with this >> kind of data set, because it lacks the multi-level upward and downward >> structure which I feel this kind of performance requires. >> >> Regards, >> >> Fergal Byrne >> >> _______________________________________________ >> 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 > >
_______________________________________________ nupic mailing list [email protected] http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org
