AI GlossaryOverfittingENTRY — Machine Learning
[Machine Learning]
Overfitting.
When a model learns training data too well, including noise, and performs poorly on new data.
In-depth explanation
01Overfitting occurs when a model becomes too complex and memorizes the training data rather than learning generalizable patterns. Signs include high training accuracy but low validation accuracy. Prevention techniques include regularization, cross-validation, early stopping, dropout, and using more training data.
Examples
02EX. 01
A decision tree that grows too deep
EX. 02
A neural network trained too long
More in Machine Learning
0301ClassificationPredicting which category or class an input belongs to from a set of predefined categories.02Cross-ValidationA technique to evaluate model performance by training and testing on different subsets of data.03Ensemble LearningCombining multiple models to produce better predictions than any single model.04FeatureAn individual measurable property or characteristic of data used as input to a machine learning model.05Feature EngineeringThe process of using domain knowledge to create new features that improve model performance.06Gradient DescentAn optimization algorithm that iteratively adjusts model parameters to minimize the loss function.
[NEXT] — APPLY THE CONCEPT
Master Overfitting.
Learn how to apply this concept with hands-on projects in our comprehensive AI programs.