Thanks, Kevin. My feeling now is that only the temperature data should be input to the NN and the tilt data should be used purely to "train" the classifier to come up with the right answer based on the measured tilt data. This would amount to a form of supervised learning and is in keeping with some other work on the same data using Echo State Networks (essentially perceptron with memory).
The problem with inputting the temperature and the tilt data is that the data set, taken as a whole, is unpredictable. The tilt is predictable given the temperature, but the data as a whole (temp+tilt) cannot be predicted from previous data (temp+tilt). It seems NuPIC does not understand cause and effect which is one of the main tools animals use to understand the world. Consider Pavlov's dogs. They learned that food would come after a bell rings. They did not try to predict WHEN the bell would ring which is a much harder, or impossible, task. On Thu, Sep 25, 2014 at 7:33 PM, Archie, Kevin <[email protected]> wrote: > John, > > I think this is an example of an important general case. On both > statistical and biological grounds I suspect you should decorrelate your > inputs before producing SDRs from them. I haven't tried this myself (have > hardly done anything with NuPIC) but I'm wondering if anyone has done > substantial preprocessing to compensate for the statistics of the inputs. > Even better would be if anyone has done a comparison of feeding NuPIC > decorrelated vs. direct sensor inputs--or a theoretical argument that > decorrelating the inputs is unnecessary or unwise. > > - Kevin > > ------------------------------ > *From:* nupic [[email protected]] on behalf of John > Blackburn [[email protected]] > *Sent:* Tuesday, September 23, 2014 7:27 AM > *To:* Archie, Kevin > *Subject:* Re: Which NuPIC tutorial do you want to see next? > > Hi Matthew, > > Rather self serving, but I would love to see a tutorial related to the > "bridge" simulation I've been trying to do or similar. I have not got NuPIC > to work yet despite some effort. Basically the difference with Hotgym is we > have 18 sensors, 10 temperature and 8 tilt (ie strain) and we want to make > predictions on all taking account of cross-correlations. So a tutorial with > multiple correlated time series would be great! > > At NPL we monitored a bridge every 5 minutes for 3 years recording all 18 > sensors so I think this data would be a great showcase for a real-world > NuPIC example. We also perturbed the bridge at known times (cutting > supports, adding weights etc) so we know when the anomalies should appear. > > John. > > On Mon, Sep 22, 2014 at 4:57 PM, Matthew Taylor <[email protected]> wrote: > >> I have more tutorials planned, but I'd like some help deciding which >> to do first. Please answer this 1-question poll: >> >> >> https://docs.google.com/forms/d/1GBYWg_-LIaYmOz9EJ5LbFo6N2ot1xv9AA22gaNdENs0/viewform?usp=send_form >> >> Thanks, >> --------- >> Matt Taylor >> OS Community Flag-Bearer >> Numenta >> >> > > ------------------------------ > > The material in this message is private and may contain Protected > Healthcare Information (PHI). If you are not the intended recipient, be > advised that any unauthorized use, disclosure, copying or the taking of any > action in reliance on the contents of this information is strictly > prohibited. If you have received this email in error, please immediately > notify the sender via telephone or return mail. >
