Overfitting

Overview

A phenomenon in machine learning where a model learns the training data too well, including its noise and outliers, resulting in poor generalization performance on unseen data. An overfit model performs well on the training set but fails to accurately predict outcomes on new, unseen data due to an overly complex model that has fit the noise, rather than generalizable patterns. This often occurs when the model is too complex relative to the amount of training data available, and is a critical issue for model development and Model Deployment.

Learning Noise

Overfitting occurs when an AI model learns training data too well, including outliers and noise in the data.

Poor Generalization

This causes poor performance on new, unseen data, which is not what you want from a model that is ready for deployment.

Controlling Complexity

To avoid overfitting, you can use techniques to control model complexity, such as regularization, and ensure that you have enough high quality training data available.

Related Concepts

• Model Training
• Model Evaluation
• Hyperparameters
• Loss Function
• Training Data