The question is therefore: is it theoretically possible to feed a neural network with random values to wait for an output since the random is lack of knowledge in the majority of cases.
For this question, I have a few examples.
First case, not a real problem?
We just throw a coin and get the result and we do it a lot of times. For each throw, we get the initial condition (air pressure, force, etc.). Now we put all this data in the neural network for it to be processed.
My guess : The result is not really random because it only depends on the initial conditions so it is possible and the neural network will do a great job. So I guess this example is not a real problem because the "degree of chance" is low.
Second case, interrogation
Now we generate a random list of numbers and phrases that match each other, so we have something like:
'zefvkbdl' -> 1613841.009 'nfeovhlzm' -> 963478.29 'jhgcjbklnsczl' -> 1.535953 'ergz' -> 9138630.26 etc ...
So that everything is generated randomly (always, the list of sentences and the number were not generated separately but each number was generated after a sentence and corresponds to this sentence). In this case, is it possible to give a neural network half the list (the list can be forever long) and expect it to predict the other half with a great precision?
My guess : It depends on the generation algorithm, but let's assume that a letter is just a particular index in an array and that the index was generated randomly. For the most part, numbers are generated randomly using time digits (the last decimal places that change extremely quickly) I am not sure – I guess it could be possible with an extremely powerful neural network theoretically do this work.
Now let's be even more theoretical and consider that it is possible to somehow store the global state of the universe at all times. The only thing that is really random to my knowledge is quantum mechanics so let's try it.
At all times, we store the state of the entire universe and the result of the measurement of a quantum particle state (like the spin of an electron). Is it possible, after the greatest formation of neural network, to "predict" the result of the measurement of a state of quantum particles knowing the state of the universe?
Since I'm just a curious student, I don't have a lot of knowledge about neural networking or quantum mechanics, so I probably said a lot of bad things and I'm sorry for that. Thank you for reading all of this and I hope someone can help me answer or correct me.
Now the real question I ask is: Does chance really exist?