Lecture 12: Monte Carlo Integration (32)

Generally what kind of distribution do we use over these areas? A normal distribution? A uniform? Seems like you'd want good coverage over everything, but there may be regions of space where the integral evals to similar values. Then there might be regions where we'd need more samples and granularity. What would be best? What kind of priors would be neccessary?

You must be enrolled in the course to comment