Module 01b
Calculus and Optimisation
Derivatives in several variables, what a gradient points at, and how the shape of a surface decides whether walking downhill on it works. Written toward optimisation, because that is what the derivatives are for here.
5 topics ~24 h estimated learning time
Sign in to keep your progress.
Ready to start
- 01b.1 Derivatives, gradients, Jacobians, Hessians
Foundation · 6 h
- 01b.2 Chain rule and computational graphs
Core · 4 h
- 01b.3 Convexity and the shape of loss surfaces
Core · 4 h
- 01b.4 Gradient descent and its variants
Core · 6 h
- 01b.5 Constrained optimization
Advanced · 4 h