Overfitting and underfitting
A model too tailored to its training data, or too simple to capture the pattern.
An overfitted model performs very well on training data and poorly on new data. An underfitted model performs poorly on both. Separate test data and cross-validation reveal the problem.
On the exam
Excellent training results with poor test results point to overfitting.
Related terms
- Training, validation and test dataSeparate datasets to build, tune and fairly measure a model.
- Model validation and testingChecking that a model is fit for its intended use before and after deployment.
AAIA Test yourself on this with timed AAIA practice questions in the AAIA Exam Simulation Hub, or read the free AAIA exam guide.
Test yourself with today’s question of the day.
