EasyDeepLearn

How is concept drift different from data drift?

medium

Answer

  • Concept drift is a change in the relationship P(yx)P(y \mid x) — the mapping from features to label evolves over time (customer behavior changes, fraud patterns evolve).
  • Data drift changes P(X); label drift changes P(y).
  • Concept drift is more dangerous because inputs may look normal but predictions become wrong.
  • Detect it by monitoring model performance on labeled data or via delayed feedback and calibration checks.
Check yourself — multiple choice
  • Concept drift = change in P(X)
  • Concept drift = change in P(yx)P(y \mid x)
  • Concept drift and data drift are the same
  • Concept drift is easy to detect without labels

Concept drift: the labeling function itself has changed.

#monitoring#data-quality

Practise MLOps & Data Quality

215 interview questions in this topic.

Related questions