Beginner · Data
Generalization curve
Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.
TL;DR. A plot showing training and validation loss over time to help detect overfitting in ML models.
Technical Definition
A plot showing training and validation loss over time to help detect overfitting in ML models.
How it works
A generalization curve illustrates how a model's performance on both training and validation datasets changes throughout the training process. By observing the trends in training loss versus validation loss, one can identify issues like overfitting, where the model performs well on training data but poorly on unseen data. This helps in diagnosing and adjusting the training strategy.
Related Concepts
- Overfitting — When a model learns noise and specific patterns in training data too well, causing it to perform poorly on new, unseen data.