Reference notes on machine learning theory, statistics and methods, written mainly for myself and kept here to revisit later.

• Bayesian networks: representation · ML Theory · Aug 2026 How directed graphs let us represent high-dimensional probability distributions compactly, and what independence assumptions a graph structure encodes.

• Exact inference by variable elimination · Methods · Aug 2026 The variable elimination algorithm for exact marginal inference, and why elimination order controls its cost.

• Maximum likelihood estimation for Bayesian networks · Statistics · Aug 2026 Why the log-likelihood decomposes across CPDs when the structure is fixed and data are fully observed, and what that buys us.