I trained a regression model with lightgbm and the learning curve doesn't look good: enter image description here

The model variance increases during training, which shows a kind of overfitting. Now, I tried many ways to fix this to no avail: regularization, features subsampling, max depth reduction, etc. Of course, the train and test datasets have been derived from the same initial dataset by a random split so they are both representative.

In your opinion, what could explain this behavior, and how could it be fixed?


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.