So let's say you have determined the optimal number of layers/neurons for your neural network. When it comes time to re-train your network with new data values (the inputs are the same, just newer data for the inputs), would the optimal number of layers/neurons need to be determined again, or would the previous numbers be sufficient for re-training?

I do not have formal training in neural networks and I don't have a background in statistics, but from the answers in the post How to set the number of neurons and layers in neural networks I get the strong indication layers/neurons describe the complexity of the relationships between the inputs/outputs and I can surmise that as long as those relationships hold in the new data, determining the optimal layers/neurons is not needed before re-training, but I'd like some more experienced opinions or facts.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.