For a machine learning algorithm prone to overfitting due to a large number of features, how does the algorithm's performance change as the dimensionality of the feature space increases?Enhanced computational efficiencyReduced overfittingImproved generalizationIncreased risk of overfitting