λ
home posts study uses projects

Learning Rate

Jul 18, 2024

machine-learning

The learning rate is a small number, usually denoted as α\alpha which is used to decrease / control the “nudge” / adjustment of the parameters of a machine learning model.

It is required because making too big adjustments might cause you to over-shoot your target. Making it too small however on the other hand will cause the model to converge too slowly.

It is one of many Hyper-Parameters in an Optimizer

LλBS

All systems up and running

Commit c0daef5, deployed on Oct 01, 2024.