Computer scienceData scienceMachine learningEnsemble learning

Adaboost

Early stopping

Report a typo

Suppose we decided to run AdaBoost for 500 iterations, and plot the test deviance (lower deviance indicates better performance):

The test set deviance curve, with the number of Adaboost iterations on the X axis and the test set deviance on the Y axis

Considering the plot above, what would be the optimal value for AdaBoost if early stopping is applied?

Select one option from the list

Around 500

Around 200

Around 300

Around 100

___

Create a free account to access the full topic