Articles

Statistical methods, bioinformatics algorithms, and computational biology — written for scientists who want both intuition and rigor.

The EM Algorithm, Part 3: The Rigorous Treatment

machine-learning

Proofs, calculus, and formal guarantees — the ELBO/KL decomposition, E-step optimality, closed-form M-step, the monotonicity theorem, and convergence to a stationary point.

The EM Algorithm, Part 2: Math with Intuition

machine-learning

The ELBO, Jensen’s inequality, KL divergence, and the E/M-step derivations — every equation explained in plain words alongside the two-coin example.

The EM Algorithm, Part 1: The Intuition

machine-learning

Understanding the Expectation-Maximization algorithm through intuition and the two-coin example — no heavy math required.