Events, Bayes, random variables, distributions and the covariance matrix that PCA and anomaly detection both need.
8h · 12 lessons · 1 challenge
Before this: Calculus & Optimization
Probability is where a loss function comes from. Cross-entropy is not a formula someone chose; it is what falls out of maximum likelihood, and this module builds to that.
It also covers the material the later modules quietly depend on: joint and conditional distributions, covariance, the covariance matrix and the multivariate Gaussian, which PCA and anomaly detection both need and which nothing else here teaches.
Ten medical-test scenarios where intuition and arithmetic disagree. Commit to a number before the answer shows.
Worth 350 XP.
