Gradient Boosting

noun

say it: “GRAY-dee-unt BOOST-ing” /ˈɡreɪdiənt ˈbustɪŋ/

A machine learning method that builds many small models in sequence, each one correcting the errors left by those before it.

What is gradient boosting?

Gradient boosting builds small decision trees one after another, each trained on the mistakes the trees before it still make, and adds up their guesses. It's often the most accurate choice for tables of data. scikit-learn's are HistGradientBoostingRegressor and HistGradientBoostingClassifier.

In a sentence

  • “Unlike a random forest, gradient boosting grows its trees one at a time, each fixing the last one's misses.”
  • “Gradient boosting lowered the energy model's error by another 8%.”

Level: Freelancer · More AI Basics terms at this level

Updated