Overfitting

Overfitting occurs when a model fits training examples or their noise too closely and performs less well on relevant new cases.

Plan for this page

Understand one concept in plain language and connect it to nearby ideas.

You will leave with: The model may memorize idiosyncrasies instead of learning patterns that generalize. Model complexity, limited data, leakage, excessive tuning, and repeated use of the same test can contribute.
Time: 3–8 min
Before you begin: No prerequisite. Keep one real example in mind.
Do this now: Read the explanation, then compare the example with the common misunderstanding.

Prepared next stepStudy generalization and testing

OverfittingOverfitting occurs when a model fits training examples or their noise too closely and performs less well on relevant new cases.

Training
Dataset
Generalization

Plain explanation

The model may memorize idiosyncrasies instead of learning patterns that generalize. Model complexity, limited data, leakage, excessive tuning, and repeated use of the same test can contribute.

Example

A plant classifier recognizes every training photo but fails when the same species appears against a different background.

Why it matters

Training success can create false confidence if independent evaluation is weak. Overfitting is one reason to separate data, document tuning, and test transfer.

Common misunderstanding

Misconception: more training always improves the model. Correction: continued optimization can improve training metrics while harming performance on new cases.

Related concepts

Follow relationships, not only the alphabet

TrainingTraining is the process of adjusting a model to improve performance on an objective using data, feedback, or interaction.DatasetA dataset is an organized collection of examples or records used for training, evaluation, analysis, or system operation.GeneralizationGeneralization is a model’s ability to perform usefully on relevant examples or conditions not seen during training.

Study generalization and testing

Sources and scope

What supports this object

Machine Learning GlossaryGoogle for Developers

Defines overfitting in machine learning; diagnosis depends on appropriate data splits, metrics and task context.

Open source

Loading page content...

…