2
$\begingroup$

Is limiting the maximum number of training epochs during optimization a standard regularization process?

I have seen it in many source codes of matrix factorization implementations, but I was not able to find it in any literature. Most publications describe more advanced ways of early stopping such as monitoring the error on validation subset.

Can somebody refer me to a scientific paper which describes this method and to which I can refer when writing my master thesis?

$\endgroup$

1 Answer 1

2
$\begingroup$

Is limiting the maximum number of training epochs during optimization a standard regularization process?

Yes, this is called early stopping. In Bishop's Pattern Recognition and Machine Learning, he develops connections between early stopping and $L^2$ regularization.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.