MIT 18.S096 Matrix Calculus For Machine Learning And Beyond, IAP 2023
5.0
(0)
16 learners
What you'll learn
This course includes
- 13.5 hours of video
- Certificate of completion
- Access on mobile and TV
Course content
1 modules • 17 lessons • 13.5 hours of video
MIT 18.S096 Matrix Calculus For Machine Learning And Beyond, IAP 2023
17 lessons
• 13.5 hours
MIT 18.S096 Matrix Calculus For Machine Learning And Beyond, IAP 2023
17 lessons
• 13.5 hours
- Lecture 1 Part 1: Introduction and Motivation57:42
- Lecture 1 Part 2: Derivatives as Linear Operators48:28
- Lecture 2 Part 1: Derivatives in Higher Dimensions: Jacobians and Matrix Functions01:13:57
- Lecture 2 Part 2: Vectorization of Matrix Functions30:18
- Lecture 3 Part 1: Kronecker Products and Jacobians53:05
- Lecture 3 Part 2: Finite-Difference Approximations51:10
- Lecture 4 Part 1: Gradients and Inner Products in Other Vector Spaces01:03:49
- Lecture 4 Part 2: Nonlinear Root Finding, Optimization, and Adjoint Gradient Methods44:26
- Lecture 5 Part 1: Derivative of Matrix Determinant and Inverse28:03
- Lecture 5 Part 2: Forward Automatic Differentiation via Dual Numbers36:02
- Lecture 5 Part 3: Differentiation on Computational Graphs32:46
- Lecture 6 Part 1: Adjoint Differentiation of ODE Solutions58:21
- Lecture 6 Part 2: Calculus of Variations and Gradients of Functionals42:32
- Lecture 7 Part 1: Derivatives of Random Functions01:06:18
- Lecture 7 Part 2: Second Derivatives, Bilinear Forms, and Hessian Matrices46:09
- Lecture 8 Part 1: Derivatives of Eigenproblems36:37
- Lecture 8 Part 2: Automatic Differentiation on Computational Graphs01:05:29
