36. Approximate Bayes for large marketing data
Covers variational inference, Laplace approximations, mini-batching, sparse models, and practical tradeoffs for large marketing datasets. You choose faster methods while checking how much accuracy they cost.