A modeling problem where a model learns the training data—including its noise or anomalies—too well, leading to poor performance on new, unseen data.
A modeling problem where a model learns the training data—including its noise or anomalies—too well, leading to poor performance on new, unseen data.