AdaBoost
Adaptive boosting ensemble for regression
AdaBoost Regressor iteratively trains base estimators, re-weighting training examples where previous models made large errors. It builds a strong regressor from simple base learners.
When to use:
- Regression with outlier-free data (AdaBoost is sensitive to noise)
- When a straightforward boosting approach is preferred
- Combining weak regressors into a competitive ensemble
Input: Tabular data with the feature columns defined during training Output: Continuous predicted value
Model Settings (set during training, used at inference)
N Estimators (default: 50) Number of boosting rounds.
Learning Rate (default: 1.0) Shrinks each estimator's contribution. Lower values improve generalization with more estimators.
Loss (default: linear)
Loss function for updating weights. linear, square, or exponential.
Base Estimator Max Depth (default: 3) Depth of the decision tree base learner.
Inference Settings
No dedicated inference-time settings.