Machine Learning

Part I: Mathematical Foundations

Every machine learning algorithm rests on three mathematical pillars: linear algebra provides the geometry of data and transformations; probability theory formalises uncertainty and inference; optimisation theory explains how models improve. This part builds each pillar from first principles, with full derivations and Python simulations.

What you will learn

āœ“Represent data as vectors and matrices and reason geometrically
āœ“Decompose matrices with eigendecomposition and SVD for compression and analysis
āœ“Model uncertainty with probability distributions and derive MLE/MAP estimators
āœ“Apply Bayes’ theorem to update beliefs as data arrives
āœ“Prove gradient descent converges on convex objectives
āœ“Derive the Adam optimiser from first principles
āœ“Formulate constrained optimisation with Lagrange multipliers and KKT conditions
āœ“Understand every Part II–VII algorithm through these three lenses
Share:XRedditLinkedIn