AIExplainer

What is gradient boosting?

A machine learning technique that combines multiple weak models to create a strong predictive model

Gradient boosting is a way to improve the accuracy of predictions by combining the predictions of many simple models. It works by training each new model to correct the mistakes of the previous models, resulting in a more accurate overall prediction.

Imagine you're trying to guess the price of a house. You ask multiple friends for their opinions, and each friend gives you a rough estimate. Gradient boosting is like taking all those estimates and combining them in a way that gives you a more accurate final guess.

A company like Netflix might use gradient boosting to predict how much a user will like a particular movie, based on their past viewing history and ratings.

Gradient boosting is commonly used in data science and machine learning for tasks such as predicting continuous outcomes, classification, and ranking. It's often used in applications like forecasting stock prices, predicting customer churn, and recommending products.

One common misconception is that gradient boosting is a single model, when in fact it's an ensemble of multiple models. Another misconception is that it's only useful for classification problems, when in fact it can be used for a wide range of tasks.

Gradient boosting was first introduced in the 1990s by researchers such as Leo Breiman and Jerome Friedman. Since then, it has become a widely used and popular machine learning technique.

boosting gradient tree boosting GBM

Three products for different needs — explore what’s relevant to you.