Overfitting
A model fitting noise and idiosyncrasies in the training data rather than the underlying pattern, so it performs well on data it has seen and worse on data it has not. It shows up as a gap between training and validation performance, not as a single bad accuracy number.
Why exams ask this
Tested with a learning-curve or train-versus-validation-score scenario. The tell is high training accuracy paired with materially lower validation accuracy. The distractor recommends increasing model capacity so it fits the training data even more closely, which makes the gap worse. The real fix is more data, regularization, or a simpler model.
Relevant to
Related concepts
Resources
No resources linked to this concept yet.