1
$\begingroup$

As we all know, XGBoost constructs trees based on gradient. I wonder how does XGBoost define gradient of MAE loss, as MAE itself is not differentiable.

After some digging of the source code, I found the implementation of MSE loss here. But I can't find any implementation of MAE loss. The original paper didn't discuss MAE loss either.

$\endgroup$

1 Answer 1

1
$\begingroup$

XGBoost does not implement MAE as a loss function.

XGBoost does implement MAE as a evaluation metric.

$\endgroup$

Your Answer

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

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