StackScholar

Classical ML

Data Leakage

Any information present at training time that will not be available at prediction time - from the target, from the future, from preprocessing fitted on the full dataset, or from entity overlap across splits.

Why interviewers ask about it

0.99 AUC on a business problem is a leakage alarm, not a success. Adversarial validation and single-feature AUC screens are the detectors to name.

Related terms

Point-in-Time Correctness

Building each training row from only the feature values that existed at that row's timestamp.

Cross-Validation

Estimating generalisation by repeatedly training on some folds and evaluating on the held-out fold.

This term is part of the free AI/ML Engineer interview preparation module - browse the full glossary for every definition.