- Article
Traditional supervisory control methods for the nonblocking control of discrete event systems often suffer from exponential computational complexity. Reinforcement learning-based approaches mitigate state explosion by sampling many random sequences i...