Learning the examples instead of the pattern.
THE DEFINITION
A model that learned its training data too specifically, including its noise, and fails on anything new.
THE ANALOGY
A student who memorised past papers rather than understanding the subject.
HOW IT SHOWS
Excellent performance on training data, poor performance on data the model has not seen.
WHY IT HAPPENS
Too little training data Too complex a model for the data available Training for too long
THE OPPOSITE PROBLEM
Underfitting: a model too simple to capture the pattern. Poor on everything.
HOW IT IS ADDRESSED
Holding back data for testing, simplifying the model, gathering more data, stopping training at the right point.
WHY YOU SHOULD KNOW IT
A vendor reporting accuracy should be reporting it on data the model never saw.
If they cannot say, ask.
RELATED TERMS
Training, model evaluation, bias.
