0
$\begingroup$

This topic confuses me. In the literature or articles, when talking about bias and variance in automatic learning, specifically in cross-validation, do they refer to the high bias (underfitting) and high variance (overfitting) in the model? Or do they refer to the bias and variance of the predictions obtained in the iterations of the cross-validation? How to handle each case?

$\endgroup$
  • 1
    $\begingroup$ I think this question is related to the „bias variance trade-off“ en.m.wikipedia.org/wiki/Bias–variance_tradeoff $\endgroup$ – Peter Jul 25 at 22:04
0
$\begingroup$

In some cases, you may have a model that's a black box - you feed input features and get output predictions, without knowing or caring what happens in the middle. In those situations, the model is, in a way, defined by its output - two models that produce the same predictions are indistinguishable from one another, even though they may be entirely distinct models. In these situations, saying a model is biased or has high variance is equivalent to saying the predictions of the model are biased or have high variance. It's acceptable to describe both the model and its output in terms of bias/variance. Cross validation provides an unbiased estimate of bias/variance for your model/predictions.

$\endgroup$
  • $\begingroup$ First.Thanks for your answer. $\endgroup$ – SRG Jun 25 at 20:14
  • $\begingroup$ Other question, in this book: books.google.com.mx/…, on page 13, the following is mentioned: "To reduce the influence of randomness introduced by data split, the k-fold cross validation can be repeated t times..." I understand that this randomness produces high variance in the results of the predictions, then, that high variance produces an model with overfitting? $\endgroup$ – SRG Jun 25 at 20:14

Your Answer

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

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