AI Blog

AI Blog

by Michele Laurelli

Learning Rate

/ˈlɜːrnɪŋ reɪt/
Training
Definition

A hyperparameter controlling how much model weights are updated during training.

Learning rate determines the step size in gradient descent. Too high causes instability, too low slows training. Common strategies: fixed, decay, cyclical, adaptive (Adam).

Examples

1

Learning rate 0.001 for Adam

2

Learning rate decay schedule

3

Warm-up then decay strategy

Michele Laurelli - AI Research & Engineering