Track 1 · Path · 7 explainers
Calculus for ML
Derivatives, the chain rule, and the optimization math behind training
- 1
Derivatives
The calculus behind gradient descent
- 2
The Chain Rule
How derivatives compose through layers
- 3
Partial Derivatives & Gradients
Extend single-variable derivatives to functions of multiple variables and gradient vectors
- 4
Logarithms & Exponentials
The math behind cross-entropy, softmax, and information theory
- 5
Matrix Calculus & Jacobians
Partial derivatives, gradients, and the Jacobian — calculus for vectors
- 6
Convexity, Minima & Saddle Points
Why some optimization problems are harder than others
- 7
Constrained Optimization
Lagrange multipliers — optimization with constraints