I posted the question also on ai.stackexchange but it didn't get any answers so I though I could try here.

Here is a copy paste:

Let's say you are training a NN in a RL setting where the state (i.e. features/input data) does not change in every step, but rather in a few (~typically around 8 steps).

For example, a state might consist of the following values:

30, 0.2, 0.5, 1, 0

And then again the same values for 6-7 times more resulting in ultimately the following input arrays:

[[30, 0.2, 0.5, 1, 0], [30, 0.2, 0.5, 1, 0], ..., [30, 0.2, 0.5, 1, 0]]

I know that the value 0 in the feature set depicts that the weight for this feature results in insignificant value.

But what about repetition of values?

How does that affect learning, if it does at all?

Any ideas?

Best, Marios.

Edit: I did not know how to search for this particular topic, it might have been answered >before, maybe you can point me there. Thanks.

Edit: I was advised to delete this question and leave it only on AI SE. However, and since I have been upvoted already, people might want to trace it back here for an answer.

Only for that reason, I leave the question here with the link that traces back to the original updated question and accepted answer.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.