I am building a neural network to predict if a video is pornographic or not by analysing the bytes of upload and download at every 0.1 seconds for a total of 25.6 seconds. So, I have 512 input variables, 256 of download and 256 of upload and I realised that for the same second, the upload and download variables are highly correlated. Does this mean that I am just adding complexity by distinguishing between upload and download variables?
Correlation between inputs of neural network does not matter. The neural network will learn which input values are associated with the target labels.
Depending on the architecture of the neural network, it will learn how different combinations of features can predict the target labels. If the high correlation is predictive, it will help the model learn.