Feature is the dimension that describes the characteristics of the sample. Regarding its interpretability in traditional machine learning and deep learning, the following statement is correct:

Question

Options

• AFeatures are interpretable in traditional machine learning, but weak in deep learning
• BFeatures are weak in traditional machine learning, but strong in deep learning
• CFeatures are weak in interpretability in traditional machine learning and deep learning
• DFeatures are interpretable in traditional machine learning and deep learning

• A
  72% (28)
• B
  3% (1)
• C
  15% (6)
• D
  10% (4)

Explanation

Option A is correct because in traditional machine learning, features are manually engineered and explicitly defined by domain experts - for example, "age," "income," or "pixel brightness" - so their meaning and contribution to a model's output can be directly examined. In deep learning, however, features are learned automatically across many layers of neural networks, producing abstract internal representations that are distributed across thousands of neurons and are difficult for humans to interpret meaningfully.

Options B and C are incorrect because they misrepresent traditional machine learning: hand-crafted, domain-specific features in traditional ML are precisely what make those models more transparent and easier to audit. Option D is incorrect because while traditional ML features are interpretable, deep learning features are not - deep networks trade interpretability for the ability to learn complex patterns from raw data.

Memory tip: Think of traditional ML as "human-designed features" (you know what they mean) versus deep learning as "machine-discovered features" (the network figures them out, but you often cannot). Interpretability follows human involvement - more human design means more interpretability.

No community discussion yet for this question.