3 ms·

So what ends up being the general purpose method for uniformly distributed random numbers in some arbitrary shape? (other than the rejection method of course)

So what ends up being the general purpose method for uniformly distributed random numbers in some arbitrary shape? (other than the rejection method of course) If arbitrary shapes always requires the rejection method, what are some "classes" that have easier algorithms (the sphere is obviously included here)

smaddox 8y ago
Not my area of expertise, but based on what I've read, you general try to get as close as possible through pure mappings, and then use rejection sampling to throw out the samples that don't fall under the desired distribution. This can work even if the distribution is dynamically changing.