What is the role of cross-validation in managing bias and variance?
Updated May 15, 2026
Short answer
Cross-validation helps estimate generalization error and identify whether a model suffers from high bias or high variance.
Deep explanation
Cross-validation splits data into multiple folds, training and validating the model repeatedly. This reduces evaluation bias and provides insight into variance by showing performance stability across folds. If performance varies widely across folds, it indicates high variance. If performance is consistently poor, it indicates high bias.
Real-world example
Used in healthcare models to ensure diagnostic accuracy is stable across patient groups.
Common mistakes
- Using only a single train-test split for evaluation.
Follow-up questions
- What is k-fold cross-validation?
- When is cross-validation not suitable?