Which phenomenon explains performance degradation when adding many irrelevant features?

The correct answer is C. Curse of dimensionality. As the number of features (dimensions) grows, the volume of the feature space increases exponentially, and training data becomes increasingly sparse relative to that space. Distance metrics lose meaning (all points appear roughly equidistant), and the model must learn boundaries

NVIDIA Certified Associate (NCA) Core AI Concepts

Question

Options

AGradient vanishing
BBias increase
CCurse of dimensionality
DUnderfitting

How the community answered

(37 responses)

B
3% (1)
C
92% (34)
D
5% (2)

Explanation

As the number of features (dimensions) grows, the volume of the feature space increases exponentially, and training data becomes increasingly sparse relative to that space. Distance metrics lose meaning (all points appear roughly equidistant), and the model must learn boundaries in a vast, poorly-sampled space - leading to degraded generalization even if training accuracy looks acceptable. Irrelevant features compound this by adding noise dimensions. Gradient vanishing (A) is a problem in deep network training dynamics. Bias increase (B) relates to underfitting, not high-dimensional noise. Underfitting (D) is a symptom, not the underlying cause; the curse of dimensionality is the mechanism.

Topics

#Curse of dimensionality#Feature engineering#Model performance#Machine learning fundamentals

Community Discussion

No community discussion yet for this question.

Full NCA-AIIO Practice