Short Answer

What test can indicate whether synthesized data is not representative?

Question: In one to three sentences, describe a way to tell whether a set of synthesized training examples is not representative, according to the source.

Sample answer: If a learning algorithm can distinguish the synthesized examples from real, non-synthesized examples—for instance because they all trace back to a narrow source like 20 car designs or 1 hour of noise—the synthesized data is likely not representative.

Key points:

  • The test is whether an algorithm can distinguish synthesized from real examples.
  • Distinguishability often stems from a narrow underlying source of synthesis.
  • Source examples include 20 car designs and 1 hour of car noise.

Rubric: Full credit: references the distinguishability test and cites at least one example from the source. Partial credit: mentions representativeness generally without the distinguishability criterion.

0

1

Updated 2026-07-10

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI