that (1) there is a bijection between state spaces and (2) through which the subMDPs have the same transition/reward