optimization
Gradient Descent
Watch a line fit itself to noisy data, one small downhill step at a time.
labs
How a handful of ML algorithms actually work, one step at a time. Press play, or step through it yourself.
Watch a line fit itself to noisy data, one small downhill step at a time.
Group points into k clusters by alternating between assigning points and recentering.
How a tree carves up a 2D dataset by repeatedly choosing the single best cut.
How activations flow through a small network, layer by layer. Visualization only — no training.