the.ai

Optimization / Schedules

verified

Learning-Rate Schedule

A good learning rate early is a bad one late. Early on the model is far from anything useful and can take large steps; near the end it needs small ones to settle. A schedule moves the rate down over the run, usually along a cosine, so the same run gets both.

Viz primitive · loss-curvesteps = 1000 · lr = 0.001
loss
step 0dashed = flat rate1000

Loss over 1000 training steps, starting near 6.0. It falls to about 1.77, with 82% of the total improvement arriving in the first half. A second line shows flat rate, ending higher at about 2.16.

Loss under a decaying schedule against a flat rate. Drag the decay to watch the schedule pull the run further down in its final stretch.

0.3

Reviewed by opendroid · 2026-08-04