Abstract: Deep Reinforcement Learning for complex control problems is often constrained by the computational cost of training, which depends heavily on the simulation environment. This paper presents ...