i have built a two-layers neural network (1000 => 1000) to predict a dynamical system driven by two real-world parameters.

When using the first parameter as input to the first layer, training the model takes 10 epochs, while the second parameters takes 1000 epochs. Does this mean that the first parameters is highly correlated with the dynamical system?

  • $\begingroup$ This could simply mean either of your input parameters for the same dynamical system contains much error or noise There's nothing unusual to take 1000 epochs for a two layer NN with 1000 units in each layer. $\endgroup$
    – cinch
    Dec 20, 2022 at 17:51
  • $\begingroup$ Im not quite understanding the specific question you are asking. Could you please provide a more detailed description of your network, your training procedures and what you are asking? $\endgroup$ Dec 21, 2022 at 8:16
  • $\begingroup$ @RobinvanHoorn It is just about the model's performance. I have an irregular time series(real-data) for which i am looking for a correlation with the two environmental parameters. $\endgroup$ Dec 21, 2022 at 12:20


You must log in to answer this question.