# What is Learning Rate Decay? Definition & Technical Mechanics

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

## Definition
Learning Rate Decay is a training hyperparameter setting that gradually decreases the optimizer's learning rate over epochs, allowing the model to make large updates early and fine adjustments later.

## Architectural Mechanics & Deep Dive
Learning rate decay is a training optimization technique where the learning rate is gradually reduced over the course of training. In early epochs, a high learning rate enables rapid exploration of parameter space; in later epochs, decaying the learning rate allows the optimizer to make fine, stable adjustments to weights, helping the model converge to a sharper minimum.

## Industry Applications & Real-World Use Cases
Model training optimization, convergence speed adjustments, and validation loss tuning.

## Frequently Asked Questions

### Why decay the learning rate?
To prevent the model from overshooting the global minimum of the cost function as training converges.

### What are common decay strategies?
Exponential decay, step decay, and cosine annealing schedules.

## Related Terminology
- [gradient-descent](https://spidits.com/ai-glossary/gradient-descent)
- [hyperparameter](https://spidits.com/ai-glossary/hyperparameter)
- [epoch](https://spidits.com/ai-glossary/epoch)

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