I've been meaning to take something like a small Scheme interpreter or whatever and run through this book with it. That said, after experimenting with PPL for

I've been meaning to take something like a small Scheme interpreter or whatever and run through this book with it. That said, after experimenting with PPL for a while, I'm more interested in variational inference, by which we can approximate the posterior distributions for models with larger hierarchical structure and higher observable dimensionality by turning the numerical integration problem into a gradient-descent problem.