Ensemble Learning

Last Updated: July 29, 2026 | By Mihail Sebastian | AI Dictionary

A machine learning approach that combines multiple models so their individual errors cancel, via three main strategies: bagging, boosting, and stacking.

What is Ensemble Learning?

Ensemble learning is a machine learning approach that combines the predictions of multiple models so their individual errors cancel, producing a combined predictor more accurate than any single member. The same idea is also called an ensemble method; the two terms are used interchangeably.

The logic is the wisdom of crowds applied to models: as long as the members are reasonably good and make different mistakes, aggregating them washes the mistakes out.

Types of Ensemble Learning

  1. Bagging trains each model independently on a random bootstrap sample of the data, then averages or votes. Random forest is the standard example: hundreds of decision trees, one vote each. Bagging mainly reduces variance.
  2. Boosting trains models in sequence, each one focused on the examples its predecessors got wrong. Gradient boosting and its fast implementation XGBoost are the standard examples. Boosting mainly reduces bias.
  3. Stacking feeds the predictions of several different models into a meta-model, which learns how to weigh and combine them into a final answer.

Bagging and boosting attack the two halves of the bias-variance tradeoff from opposite ends, which is why the choice between them depends on how the base model fails.

Example of Ensemble Learning

The Netflix Prize, a public competition to improve the accuracy of Netflix’s movie-rating predictions by 10%, was won in 2009 by a team whose solution was an ensemble: a blend of hundreds of individual models, each weak in a different way, combined into one predictor that no single model could match.

The same pattern runs a spam filter. One classifier reads word frequencies, another sender reputation, a third message structure. Each alone misfires on some emails; their combined vote catches spam that any single model would let through, with fewer false alarms on legitimate mail.

Related AI terms: Ensemble Method · Random Forest · XGBoost · Decision Tree · Bias-Variance Tradeoff

Did you like the Ensemble Learning gist?

Learn about 250+ need-to-know artificial intelligence terms in the AI Dictionary.

Mihail Sebastian — Writes about AI governance, regulation, and the technology behind them. Placeholder bio — replace with a real credential line. About

Read the Governor's Letter

Stay ahead with Governor's Letter, the newsletter delivering expert insights, AI updates, and curated knowledge directly to your inbox.

By subscribing to the Governor's Letter, you consent to receive emails from AI Guv.
We respect your privacy - read our Privacy Policy to learn how we protect your information.

Browse All AI Terms A–Z

Every term in the dictionary, in alphabetical order. Jump to a letter or scroll the full list.

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W

X

Y

Z