Outliers can be bad for boosting because boosting builds each tree on previous trees' residuals/errors. Outliers will have much larger residuals than non-outliers, so gradient boosting will focus a disproportionate amount of its attention on those points.
referenceS:
https://stats.stackexchange.com/questions/140215/why-boosting-method-is-sensitive-to-outliers
No comments:
Post a Comment