Quote:
|
Originally Posted by Structural Integrity
Hmmmm, I thought they already used the data collected over the past 100 years to predict the weather.
Something along the lines of "Oh, high and low pressure here and there, tempuratures like that, about the same as on the 20th of Februari in 1990, and it rained the day after, so we predict rain for tomorrow".
Roughly said...
Ok, it's not learning, but comparing.
|
Yes, you have a finite set of data and youre trying to extrapolate rules from it to allow future events to be correctly classified. This is all that learning means, in an AI/statistics context. Machine "learning" here is just any method that allows a computer to go beyond the data it has available, and predict new cases.
Quote:
|
But anyway, I don't think this learning thing exists because of the huge amounts of data that needs to be processed, and the lack thereof in certain areas.
|
It does exist, otherwise our weather reports would be competely random. The rules which have been learnt arent perfect, but that is due to the complexity of the data set.
Quote:
|
Even if you knew the weather conditions of every place on the planet, their interactions, how they influence eachother, are in my opinion too complex to be defined by a set of finite rules.
|
If you literally knew everything then you could compute the weather everywhere for the next 10 billion years, assuming that you dont have any fundamental physical indeterminacy (eg quantum nonsense, human 'freewill' affecting conditions, whatever).
edit: although if you just knew the weather conditions on this planet then you wouldnt know everything, so your predictions could still be screwed up by external influences like meteors flying past earth.