CompTIA DataX DY0-001 (V1) Practice Question

The correct answer is Bootstrap Aggregation (Bagging). The scenario describes a model with high variance, as indicated by the large gap between its performance on the training data (99%) and the test data (75%). Bagging is an ensemble technique specifically designed to reduce variance. It works by creating multiple bootstrap samples (random samples with replacement) from the training data, training a separate model on each sample, and then aggregating their predictions (e.g., by voting or averaging). This process smooths out the predictions and makes the final model more robust and less prone to overfitting.

Gradient Boosting is incorrect because it is an ensemble method that primarily aims to reduce bias, not variance. It builds models sequentially, with each new model attempting to correct the errors of the previous ones. Applying boosting to an already overfitted model could worsen the problem.

Stacking is an ensemble method where a new model (a meta-learner) is trained to combine the predictions of several different base models (heterogeneous learners). While it is a powerful technique for improving predictions, its primary goal is to find the optimal combination of diverse models, not specifically to reduce the variance of a single, overfitted model type.

Adversarial Training is incorrect because it is a technique used to improve a model's robustness against adversarial examples, which are inputs intentionally crafted to cause the model to make a mistake. It is not a standard ensemble learning method for mitigating general overfitting caused by model complexity.