my tutorial: "machine learning and Bayesian inference"

[gusl]

I've posted the slides for today's tutorial.

I feel unsure about my usage of "Bayesian Model Averaging", since it's actually averaging over the parameter in an IID Bernoulli model. But I guess it depends on what you call a model: if Ber(p1) is considered a "different model" from Ber(p2), then my usage is justified.