Reinforcement learning for anisotropic p-adaptation and error estimation in high-order solvers | Publicación