Bias-variance tradeoff
The relationship between a model too simple to capture the real pattern, which underfits and has high bias, and a model that memorizes the training data too closely, which overfits and has high variance. Reducing one usually increases the other, which is why model complexity is tuned rather than maximized.
Why exams ask this
Tested as matching a symptom to a cause: high error on both training and validation data points to bias, high training accuracy with poor validation accuracy points to variance. The exam wants the correct fix paired to the correct diagnosis, not one generic answer for both.
Relevant to
Related concepts
Resources
No resources linked to this concept yet.