# Optimization Practice

> ML preparation set.

- Stable ID: `optimization-practice`
- Area: machine-learning
- Kind: practice
- Timebox: 30 minutes

Relate stationary points, convexity, objectives, regularization, and training diagnostics.

## Instructions

1. Separate objective geometry from optimizer behavior.

## Ordered items

1. [Convexity and stationary points](https://mlprep.iwase.dev/machine-learning/optimization/original-ml-optimization/) — `original-ml-optimization` (8 min)
2. [L1 versus L2 regularization](https://mlprep.iwase.dev/machine-learning/optimization/original-ml-regularization/) — `original-ml-regularization` (6 min)
3. [Gradient descent on a quadratic](https://mlprep.iwase.dev/machine-learning/gradient-descent/original-ml-gd/) — `original-ml-gd` (7 min)

## Completion

Diagnose whether a problem comes from the objective, step size, or model complexity.