Underfitting
Underfitting happens when a model is too simple or too poorly trained to capture the patterns in its data, so it performs badly even on the training set.
What is Underfitting?
An underfitted model misses real relationships, for example fitting a straight line to clearly curved data. It is the opposite problem to overfitting, and good model development seeks a balance between the two.
How it works
Remedies include using a more expressive model, adding informative features, training for longer or reducing excessive regularisation.
← All Artificial Intelligence & Superintelligence definitions · ← Back to the A–Z definitions