|View source on GitHub|
Returns a program that takes a Metropolis step with an inner kernel.
The Metropolis algorithm is a special case of Metropolis-Hastings for
symmetric proposal distributions. This algorithm assumes the
program is symmetric (i.e. p(y | x) = p(x | y)).
unnormalized_log_prob: A function that computes the log probability of a
inner_step: A probabilistic program that acts as the proposal distribution
for a Metropolis step.
A program that proposes a new state and accepts or rejects according to the
unnormalized log probability.