In the context of ensemble learning, what is the term for sampling the training data with replacement to create multiple diverse models?
Show answer
Bootstrapping
Bootstrapping is the 'B' in Bagging (Bootstrap Aggregating). It allows for the training of multiple models (like trees in a Random Forest) on slightly different subsets of data, which helps reduce the variance of the final ensemble.
Also asked as:
- What term describes the process of sampling training data with replacement to generate diverse models for an ensemble?