Drag to orbit. Blue/amber beads: data above/below the plane. Steel ball: current weights on the loss basin.

Data Science: A Physics Perspective

Linear regression trained by gradient descent with momentum, seen as a steel ball rolling down a loss basin next to the data cloud it is fitting. The step size is measured against the exact stability limit, the correlation between features stretches the valley, and seven independent numerical checks confirm the maths live.

Made with 🟠 Blender · Three.js