Class track
Mathematical Foundations for Machine Learning
Intuition → geometry → mathematics → code → application.
1. Matrices as transformations
See matrices as actions on space, not just tables of numbers.
2. Basis, span, rank
Coordinate systems, independence, and information preservation.
3. Eigenvectors and eigenvalues
Stable directions and why they matter.
4. PCA and SVD
Structure, compression, dimensionality reduction.
5. Probability and Bayes
Reasoning under uncertainty.
6. Regression and likelihood
Prediction, loss, and statistical fitting.
7. Gradient descent and Adam
How optimization actually moves parameters.
8. Neural networks from scratch
Build the core mechanics without hiding behind frameworks.