Reinforced Model Predictive Control via Trust-Region Quasi-Newton Policy Optimization | Read Paper on Bytez