# What is Early Stopping? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Model Training  
> **Canonical Citation:** [https://spidits.com/ai-glossary/early-stopping](https://spidits.com/ai-glossary/early-stopping)

## Definition
Early Stopping is a regularization technique that halts a model's training process when its performance on a separate validation dataset stops improving, even if the training loss continues to decrease.

## Architectural Mechanics & Deep Dive
Early stopping is a regularization technique that halts the training process before the specified number of training epochs is completed. It monitors the model's performance on an independent validation dataset; once the validation loss stops improving and begins to rise—indicating that the model is starting to overfit the training data—training is terminated, and the best-performing weights are restored.

## Industry Applications & Real-World Use Cases
Training optimization, overfitting prevention, and resource conservation

## Frequently Asked Questions

### How does early stopping work?
It monitors validation loss epoch-by-epoch. If validation loss plateaus or starts rising for a set number of epochs (patience), training is aborted.

### What is the risk of stopping training too early?
Underfitting, where the model has not had enough iterations to capture the core patterns in the training set.

## Related Terminology
- [overfitting](https://spidits.com/ai-glossary/overfitting)
- [epoch](https://spidits.com/ai-glossary/epoch)
- [bias-variance-tradeoff](https://spidits.com/ai-glossary/bias-variance-tradeoff)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*