● mastered · kind concept · level 4 · 20h

The gradient, Jacobian, and Hessian generalise the derivative to many dimensions. The gradient points uphill on the loss surface; optimisation walks the opposite way.

Enlaces