Skip to content
Premium Content

Matrix and Vector Calculus - The Machinery of Machine Learning

Derivatives when inputs and outputs are vectors and matrices: gradients, Jacobians, and Hessians that power every optimization and neural network

This chapter requires a subscription to access.

What you'll unlock:

  • 1. The Gradient Vector: Derivative of a Scalar by a Vector
  • 2. The Jacobian Matrix: Derivative of a Vector by a Vector
  • 3. Gradients of Matrix Expressions: The Matrix Cookbook Essentials
  • 4. The Vector and Matrix Chain Rule
  • 5. The Hessian: Curvature of Scalar Fields
  • 6. Least Squares by Matrix Calculus: The Normal Equations
  • 7. Matrix Calculus in Machine Learning: A Linear Layer End to End
Subscribe to Unlock

Already have an account? Sign in