Key points are not available for this paper at this time.
Safe Reinforcement Learning can be defined as the process of learning policies that maximize the expectation of the return in problems in which it is important to ensure reasonable system performan...
GarcíaJavier et al. (Thu,) studied this question.