nerdexam
Huawei

H13-311_V3.5 · Question #126

The training error will continue to decrease as the model complexity increases.

The correct answer is A. TRUE. A (TRUE) is correct because training error measures how well a model fits the data it was trained on, and a more complex model always has at least as much capacity to fit that data as a simpler one - in the limit, a sufficiently complex model can memorize every training…

Machine Learning Basics

Question

The training error will continue to decrease as the model complexity increases.

Options

  • ATRUE
  • BFALSE

How the community answered

(28 responses)
  • A
    79% (22)
  • B
    21% (6)

Explanation

A (TRUE) is correct because training error measures how well a model fits the data it was trained on, and a more complex model always has at least as much capacity to fit that data as a simpler one - in the limit, a sufficiently complex model can memorize every training example, driving training error to zero. This is the core mechanism behind overfitting: the model learns the noise and idiosyncrasies of the training set rather than the underlying pattern.

B (FALSE) is wrong because it conflates training error with generalization (test) error. While test error does eventually rise with excess complexity (the classic U-shaped bias-variance curve), training error is monotonically non-increasing as complexity grows - it never goes back up.

Memory tip: Think of a student who memorizes the answer key. The more study time (complexity) they get, the closer their score on that exact test (training error) gets to 100% - but they may still fail a new exam (test error). Training error = score on the answer key you already have.

Topics

#model complexity#training error#bias-variance tradeoff#overfitting

Community Discussion

No community discussion yet for this question.

Full H13-311_V3.5 Practice