Loading…
Glossary term
When a model is too simple to capture the patterns in the data and performs poorly even on training data.
An underfit model is too simple, or trained too little, to capture the real pattern. It does poorly on both the training data and new data.
Fixes include a more flexible model, better features and training for longer.
Underfitting and overfitting are the two ends of the central trade-off in machine learning: finding a model that is just flexible enough.