AI & Machine LearningDeep Learning
How a Machine Learns
A model improves by following tiny slopes toward smaller mistakes
A model starts out wrong about everything. Then it gets better, one tiny nudge at a time. Nudged which way, though?
You'll find out
- How a machine measures its mistakes
- Why the step size matters so much
- What training repeats over and over
Why it matters
Every AI you use was trained by this loop. Seeing it once makes the word training much less mysterious.
Play this lesson
Free to play in your browser or on iPhone