Glossary term
Glossary term
Training and Fine-Tuning
A machine learning technique that iteratively combines a set of simple and not very accurate classification models (referred to as "weak classifiers") into a classification model with high accuracy (a "strong classifier") by upweighting the examples that the model is currently misclassifying.
See Gradient Boosted Decision Trees? in the Decision Forests course for more information.
Created for this library
A credit risk team uses boosting to combine many shallow trees into a strong scorecard for personal loans.
A retail analytics team uses gradient boosting to predict next-week sales by SKU because boosting handles tabular features and missing data well.
A search ranking team uses boosting on hand-crafted features as a strong baseline before evaluating neural rankers.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License