EasyDeepLearn

What criterion does a regression tree use to choose splits?

easy

Answer

  • Variance reduction: pick the split that most reduces sum of squared errors of the target within child nodes.
  • Equivalently, maximize the total variance of the parent minus the weighted sum of children variances.
  • The prediction in each leaf is the mean of the target for training points that fall into it (median for MAE-loss trees).
  • Same tree machinery, different impurity function.
Check yourself — multiple choice
  • They use Gini impurity
  • Variance reduction (MSE) — predict the leaf mean of the target
  • They use entropy of y
  • They use cross-entropy

Regression trees minimize squared error (or MAE) — leaf predictions are the target mean.

#decision-trees#trees#metrics-regression

Practise Supervised Learning

215 interview questions in this topic.

Related questions