generalization

The ability of an AI model to perform well on unseen data, beyond the examples it was trained on, indicating its robustness and learning efficacy.

242 papers