AlgoPlus//machine learning / random-forest
Read the theory

Random Forest

Many decorrelated trees vote; the majority is far more robust than any single tree.

Trees voted
Prediction
Your labelled points (x,y,class — class 0 or 1)
Legend
Class 1
Class 0
Query
AI Tutor Workspace
In a nutshell
A random forest is an ensemble — a whole team of decision trees that vote. Each tree is trained on its own random sample of the data (bagging) and considers only a random subset of features at each split, which stops the trees from all making the same mistakes. For a new point, every tree predicts and the majority wins, giving an answer far steadier than any single tree.
Ready
Press play to begin the cinematic walkthrough.
Ask a crowd of slightly-different experts and go with the majority. Because each tree saw different data, their mistakes don't line up, so the vote is steadier than any one of them.
Key terms
Go deeper in the lesson
Read the full theory, intuition & complexity for Random Forest.