Thanks, I will follow through the tutorial you posted and then try again. I
will post the swarm description then.

What I want it to do is primarily predict tilt(t+1) from temperature(t), I
do not want it to predict temperature(t+1) from temperature(t) as this will
not work well with the data. (unlike Hotgym, the temperature cannot be
predicted from previous temperatures)

If I set the predicted field as "tilt" will it refrain from trying to
predict the temperature also? I want to avoid it trying to predict
temperature. Basically I want to shut down some of the prediction
combinations in favour of the useful ones. I could imagine doing this
manually (but don't know how to do it) but the best would be if the HTM can
figure out the useful combinations itself. I think you are saying swarming
does this? I will see if it works.

John.

On Mon, Sep 29, 2014 at 6:24 PM, Subutai Ahmad <[email protected]> wrote:

> Hi John,
>
> I don't recall if you were swarming over the data. If so, did you set tilt
> as the predicted field in the swarm?  If so, it should always use past
> values of tilt to predict future values. Additionally, if temperature
> provides incremental information over and above past values of tilt, it
> should also include that in the model.  Maybe you could post your swarm
> parameters here and someone could help you debug.
>
> Thanks,
>
> --Subutai
>
> On Mon, Sep 29, 2014 at 10:04 AM, John Blackburn <
> [email protected]> wrote:
>
>> Subatai,
>>
>> I think the problem is, we are feeding temperature and tilt and NuPIC is
>> saying "temperature and tilt are strongly correlated, so I'll forget the
>> tilt and just try to predict temperature". The trouble is temperature is a
>> weather effect and fundamentally unpredictable. It's hotter in the day and
>> during the summer but the precise details are unpredictable. That is, the
>> temperature data is not nearly as cyclic as the Hotgym example etc.
>>
>> We need some way of telling Nupic "you don't need to predict temperature,
>> you need to predict tilt GIVEN the temperature". You might argue this is
>> the sort of task a traditional stateless perceptron network would do
>> better, but the best result so far has been achieved with an echo state
>> network (ESN). The ESN had 10 temperature inputs, 8 tilt outputs and was
>> trained to give the right outputs (supervised learning). Between the input
>> and outputs is a "reservoir" of randomly connected recurrent neurons which
>> maintains a short term memory. Thus the ESN has a sense of time and
>> sequence like the HTM. But it also has the concept of "cause and effect"
>> which I think HTM lacks (or please correct me).
>>
>> Analogously, Pavlov's dogs were able to predict that food would come
>> after a bell rang. The ringing of the bell was unpredictable but the food
>> was predictable given the ringing of the bell. This is analogous to the
>> bridge with bell = temperature, food = tilt. Can HTM "reason" like the dogs
>> did?
>>
>> On Fri, Sep 26, 2014 at 9:35 PM, Subutai Ahmad <[email protected]>
>> wrote:
>>
>>>
>>> >if you have a problem with big stationary correlational structure in
>>> the inputs--
>>> >you should transform it out so NuPIC can work on the non
>>> obvious features
>>>
>>> Agreed. In some sense this is what pooling does.  The spatial pooler
>>> transforms spatially correlated (but very different) patterns into a
>>> consistent SDR pattern.  The temporal pooler transforms temporally
>>> predictable (but potentially totally different) SDR patterns into a single
>>> coherent SDR representation.  The encoders also can help greatly with this.
>>>
>>> --Subutai
>>>
>>> On Fri, Sep 26, 2014 at 12:01 AM, Archie, Kevin <[email protected]>
>>> wrote:
>>>
>>>>  Subutai,
>>>>
>>>>  Thanks for the link, interesting reading. I'm amused by the problem
>>>> of it being (1) a problem NuPIC isn't really well suited to and (2) the
>>>> first thing lots of people are going to try anyway.
>>>>
>>>>  Thinking about statistics, it seems to me that if you have a problem
>>>> with big stationary correlational structure in the inputs--either temporal,
>>>> as the sine wave example, or spatial, as John's bridge data--you should
>>>> transform it out so NuPIC can work on the nonobvious features. Thinking
>>>> about biology, by the time you get to cortex a lot of the stationary
>>>> structure has been filtered out -- think about the processing in the
>>>> retina. It's possible that NuPIC is good enough to solve lots of
>>>> interesting problems even without preprocessing, but I suspect that some
>>>> care to the input representation could greatly help with performance (in
>>>> cycles, at least by reducing the input dimensionality) and performance (in
>>>> error rate, by getting rid of lots of chaff).
>>>>
>>>>  Or I could be all wet. It happens.
>>>>
>>>>    - k
>>>>
>>>>  On Sep 25, 2014, at 1:53 PM, Subutai Ahmad wrote:
>>>>
>>>>  Hi Kevin,
>>>>
>>>>  I did some simple experiments with swarming and correlated inputs
>>>> [1]. One thing to note is that temporal correlation / sequence structure is
>>>> also very important. That is independent from spatial correlation.
>>>>
>>>>  --Subutai
>>>>
>>>>  [1]
>>>> https://github.com/subutai/nupic.subutai/tree/master/swarm_examples
>>>>
>>>> On Thu, Sep 25, 2014 at 11:33 AM, 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.
>>>>>
>>>>
>>>>
>>>>
>>>
>>
>

Reply via email to