---
id: 2023-12-17
aliases: December 17, 2023
tags:
- link-note 
- Data-Science
- Machine-Learning 
- Gradient-descent 
---
# Gradient Descent

- Update parameters that minimize values of loss functions 
- **Instantaneous is always _0_**
- Therefore, update parameter that a derivative of loss function is equal to 0

## Pseudo Code

1. Find the derivative of the loss function at the current parameters. 
2. Update parameters in the opposite direction of the derivative 
3. Repeat steps 1 and 2 as many epochs (hyperparameter) until the differential value becomes 0.