RP, or Randomized Polynomial time, is a complexity class used in computational complexity theory that describes problems for which a solution can be verified in polynomial time with a probabilistic Turing machine that has a bounded error probability. Specifically, if the answer is 'yes', the algorithm will produce a correct result with a probability of more than 1/2, and if the answer is 'no', it will always produce the correct result.